2024-03-28T18:18:57Zhttps://escholarship.org/oaioai:escholarship.org:ark:/13030/qt4tp227hz2023-05-04T01:25:14Zqt4tp227hzThe Capacity of Classical Summation over a Quantum MAC with Arbitrarily Replicated InputsYao, YuhangJafar, Syed A2023-05-03application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/4tp227hzpublicationoai:escholarship.org:ark:/13030/qt44p655jp2023-04-18T01:41:25Zqt44p655jpN-Sum Box: An Abstraction for Linear Computation over Many-to-one Quantum NetworksAllaix, MatteoLu, YuxiangYao, YuhangPllaha, TefjolHollanti, CamillaJafar, Syed A2023-04-17application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/44p655jppublicationoai:escholarship.org:ark:/13030/qt0s5031z62021-05-09T18:55:55Zqt0s5031z6On the Synergistic Benefits of Reconfigurable Antennas and Partial Channel Knowledge for the MIMO Interference ChannelYuan, BofengIsmailoglu, NilabJafar, Syed A2021-05-09Blind Interference Alignment (BIA) schemes create and exploit channel coherence patterns without the knowledge of channel realizations at transmitters, while beamforming schemes rely primarily on channel knowledge available to the transmit- ters without regard to channel coherence patterns. In order to explore the compatibility of these disparate ideas and the possibility of synergistic gains, this work studies the Degrees of Freedom (DoF) of the 2-user (?1 × ?1)(?2 × ?2) Multiple- Input Multiple-Output (MIMO) Interference Channel (IC) where Transmitter 1 is equipped with reconfigurable antennas and has no channel knowledge, while Transmitter 2 has partial channel knowledge but no reconfigurable antennas. Taking a fundamental dimensional analysis perspective, the main question is to identify which antenna configurations allow synergistic DoF gains. The main results of this work are two-fold. The first result identifies antenna configurations where both reconfigurable antennas and partial channel knowledge are individually beneficial, as those where ?1 < ?1 < min(?2,?2). The second result shows that synergistic gains exist in each of these settings, over the best known solutions that rely on either reconfigurable antennas or partial channel knowledge alone. Coding schemes that jointly exploit partial channel knowledge and reconfigurable antennas emerge as a byproduct of the analysis.Blind Interference AlignmentDegrees of Free- domInterference ChannelMIMOPartial CSITapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/0s5031z6publicationoai:escholarship.org:ark:/13030/qt8nh0m0qm2020-11-15T15:50:58Zqt8nh0m0qmExploring Aligned-Images Bounds:Robust Secure GDoF of 3-to-1 Interference ChannelChan, Yao-ChiaJafar, Syed A2020-11-15Sum-set inequalities based on Aligned-Images bounds have been recently introduced as essential elements of converse proofs for asymptotic/approximate wireless network capacity characterizations under robust assumptions, i.e., as- sumptions that limit channel knowledge at the transmitters to finite precision. While these sum-set inequalities have produced robust Generalized Degrees of Freedom (GDoF) results for various wireless networks, their scope and limitations in general are not well understood. To explore these limitations, in this work we study the robust secure GDoF of a symmetric 3-user many- to-one interference channel. We identify regimes where existing sum-set inequalities are sufficient, settling the GDoF for those settings. For the remaining regime we conjecture the form of new sum-set inequalities that may be needed, whose validity remains an open problem for future work.SecurityFinite Precision CSITAligned Images BoundsSumset Inequalitiesapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/8nh0m0qmpublicationoai:escholarship.org:ark:/13030/qt8fx6x83p2019-10-31T17:57:42Zqt8fx6x83pGeneralized Cross Subspace Alignment Codes for Coded Distributed Batch Matrix MultiplicationJia, ZhuqingJafar, Syed A2019-10-31The goal of coded distributed batch matrix multiplication is to efficiently multiply L instances of \lambda x \kappa matrices, A=(A_1, ..., A_L)$, with L instances of \kappa x \mu matrices B=(B_1,..., B_L), by distributing the computation across S servers, such that the response from any R servers (R is called the recovery threshold) is sufficient to compute the L matrix products, AB=(A_1B_1, A_2B_2, ..., A_LB_L). Existing solutions either compute each $A_lB_l$ one at a time by partitioning individual matrices and coding across these partitions, or rely only on batch processing, i.e., coding across the batch of matrices without any matrix partitioning. The state-of-art for matrix-partitioning and batch processing approaches is represented by Entangled Polynomial Codes (EP codes), and Lagrange Coded Computing (LCC), respectively. In order to combine the benefits of the two approaches, we propose Generalized Cross-Subspace Alignment Codes (GCSA codes) that unify, generalize and improve upon the state of art. GCSA codes bridge the two extremes by efficiently combining both matrix-partitioning and batch processing, and offer flexibility in how much of each approach is used. Both EP codes and LCC codes can be recovered as special cases of GCSA codes. Remarkably, even without matrix partitioning, GCSA codes demonstrate an advantage over LCC codes in download-constrained settings. This is due to cross-subspace alignment, characterized by a Cauchy-Vandermonde code structure that aligns interference along Vandermonde terms, while the desired matrix products remain resolvable along Cauchy terms.Distributed Matrix MultiplicationCross Subspace AlignmentInterference AlignmentDistributed Computingapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/8fx6x83ppublicationoai:escholarship.org:ark:/13030/qt4242x6082019-10-27T23:34:29Zqt4242x608Robust Optimality of TIN under Secrecy ConstraintsChan, Yao-chiaGeng, ChunhuaJafar, Syed A2019-10-27A parameter regime is identified where the simple scheme of treating interference as Gaussian noise (TIN), with power control and jamming, is optimal for the secure generalized degrees of freedom (GDoF) region of Gaussian broadcast networks under the robust assumption of finite-precision channel state information at the transmitter (CSIT). The network consists of one transmitter equipped with K antennas, and K single-antenna receivers. The results are generalized to groupcast (equivalently, compound broadcast) settings where each message is desired by a disjoint group of receivers. Noting that messages are independently encoded in the GDoF-optimal scheme, the result for the broadcast channel is extended to its counterpart Gaussian interference channel under finite precision CSIT. Evidently, both secrecy constraints and finite precision CSIT limit the benefits of more sophisticated schemes, leading to optimality of simpler schemes for larger parameter regimes. Aligned Image bounds are the key to the proof of optimality for these larger parameter regimes under finite precision CSIT.Treating Interference as Noise (TIN)Generalized Degrees of Freedom (GDoF)SecurityBroadcastInterference Channelapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/4242x608publicationoai:escholarship.org:ark:/13030/qt5gt0b9v12019-08-27T17:21:16Zqt5gt0b9v1$X$-secure $T$-private Information Retrieval from MDS Coded Storage with Byzantine and Unresponsive ServersJia, ZhuqingJafar, Syed A2019-08-27The problem of $X$-secure $T$-private information retrieval from MDS coded storage is studied in this paper, where the user wishes to privately retrieve one out of $K$ independent messages that are distributed over $N$ servers according to an MDS code. It is guaranteed that any group of up to $X$ colluding servers learn nothing about the messages and that any group of up to $T$ colluding servers learn nothing about the identity of desired message. A lower bound of achievable rates is proved by presenting a novel scheme based on \emph{cross-subspace alignment} and a successive decoding with interference cancellation strategy. For large number of messages $(K\rightarrow\infty)$ the achieved rate, which we conjecture to be optimal, improves upon the best known rates previously reported in the literature by Raviv and Karpuk, and generalizes an achievable rate for MDS-TPIR previously found by Freij-Hollanti et al. that is also conjectured to be asymptotically optimal. The setting is then expanded to allow unresponsive and Byzantine servers. Finally, the scheme is applied to find a new lower convex hull of (download, upload) pairs of secure and private distributed matrix multiplication that generalizes, and in certain asymptotic settings strictly improves upon the best known previous results.Private Information RetrievalPrivate Secure Distributed Matrix MultiplicationCross-Subspace AlignmentSuccessive DecodingInterference Cancellationapplication/pdfCC-BY-NC-SAeScholarship, University of Californiahttps://escholarship.org/uc/item/5gt0b9v1publicationoai:escholarship.org:ark:/13030/qt0mf232ft2019-05-10T23:45:48Zqt0mf232ftGDoF of Interference Channel with Limited Cooperation under Finite Precision CSITWang, JungeYuan, BofengHuang, LexiangJafar, Syed A2019-05-10The Generalized Degrees of Freedom (GDoF) of the two user interference channel are characterized for all parameter regimes under the assumption of finite precision channel state information at the transmitters (CSIT), when a limited amount of cooperation is allowed between the transmitters in the form of π DoF of shared messages. In all cases, the number of over- the-air bits that each cooperation bit buys is shown to be equal to either 0, 1, 1/2 or 1/3.GDoFCooperationInterference ChannelBroadcast ChannelAligned Image SetsFinite Precision CSITapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/0mf232ftpublicationoai:escholarship.org:ark:/13030/qt0qn447j42019-01-20T19:00:34Zqt0qn447j4Towards an Extremal Network Theory – Robust GDoF Gain of Transmitter Cooperation over TINChan, Yao-ChiaJafar, Syed A2019-01-20With the emergence of aligned images bounds, significant progress has been made in the understanding of robust fundamental limits of wireless networks through Generalized Degrees of Freedom (GDoF) characterizations under the assumption of finite precision channel state information at the transmitters (CSIT), especially for smaller or highly symmetric network settings. A critical barrier in extending these insights to larger and asymmetric networks is the inherent combinatorial complexity of such networks. Motivated by other fields such as extremal combinatorics and extremal graph theory, we explore the possibility of an extremal network theory, i.e., a study of extremal networks within particular regimes of interest. As our test application, we study the GDoF benefits of transmitter cooperation over the simple scheme of power control and treating interference as Gaussian noise (TIN) for three regimes of interest where the interference is weak. The question is intriguing because while in general transmitter cooperation can be quite powerful, finite precision CSIT and weak interference favor TIN. The three regimes that we explore include a TIN regime previously identified by Geng et al. where TIN was shown to be GDoF optimal for the $K$ user interference channel, a CTIN regime previously identified by Yi and Caire where the GDoF region achievable by TIN turns out to be convex without the need for time-sharing, and an SLS regime previously identified by Davoodi and Jafar where a simple layered superposition (SLS) scheme is shown to be optimal in the $K$ user MISO BC, although only for $K\leq 3$. It remains an intriguing possibility that TIN may not be far from optimal in the CTIN regime, and that SLS schemes may be close to optimal even for larger networks in the SLS regime, but the curse of dimensionality is one of the obstacles that stands in the way of such generalizations. Under finite precision CSIT, appealing to extremal network theory we obtain the following results. In the TIN regime as well as the CTIN regime, we show that the extremal GDoF gain from transmitter cooperation over TIN is $\Theta(1)$, i.e., it is bounded above by a constant regardless of the number of users $K$. In fact, the gain is at most a factor of $2$ in the CTIN regime, which automatically implies that TIN is GDoF optimal within a factor of $2$ in the CTIN regime. In the TIN regime, the extremal GDoF gain of transmitter cooperation over TIN is shown to be exactly $50\%$, regardless of the number of users $K$, provided $K>1$. However, in the SLS regime, the extremal GDoF gain of transmitter cooperation over TIN is $\Theta(\log_2(K))$, i.e., it scales logarithmically with the number of users. Remarkably, an SLS scheme suffices to demonstrate this extremal GDoF gain. Last but not the least, as a byproduct of our analysis we prove a useful cyclic decomposition property of the sum GDoF achievable by TIN in the SLS regime.Generalized Degrees of Freedom (GDoF)Channel State information at Transmitters (CSIT)Treating Interference as Noise (TIN)application/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/0qn447j4publicationoai:escholarship.org:ark:/13030/qt0qh0h88p2018-12-09T07:35:15Zqt0qh0h88pOn the Capacity of Locally Decodable CodesSun, HuaJafar, Syed A2018-12-08A locally decodable code (LDC) maps $K$ source symbols, each of size $L_w$ bits, to $M$ coded symbols, each of size $L_x$ bits, such that each source symbol can be decoded from $N \leq M$ coded symbols. A perfectly smooth LDC further requires that each coded symbol is uniformly accessed when we decode any one of the messages. The ratio $L_w/L_x$ is called the symbol rate of an LDC. The highest possible symbol rate for a class of LDCs is called the capacity of that class. It is shown that given $K, N$, the maximum value of capacity of perfectly smooth LDCs, maximized over all code lengths $M$, is $C^*=N\left(1+1/N+1/N^2+\cdots+1/N^{K-1}\right)^{-1}$. Furthermore, given $K, N$, the minimum code length $M$ for which the capacity of a perfectly smooth LDC is $C^*$ is shown to be $M = N^K$. Both of these results generalize to a broader class of LDCs, called universal LDCs. The results are then translated into the context of PIR$_{\max}$, i.e., Private Information Retrieval subject to maximum (rather than average) download cost metric. It is shown that the minimum upload cost of capacity achieving PIR$_{\max}$ schemes is $(K-1)\log N$. The results also generalize to a variation of the PIR problem, known as Repudiative Information Retrieval (RIR).Private Information RetrievalLocally Decodable CodesCapacityapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/0qh0h88ppublicationoai:escholarship.org:ark:/13030/qt70d1v0jt2018-10-24T20:10:47Zqt70d1v0jtOn the Necessity of Non-Shannon Information Inequalities for Storage Overhead Constrained PIR and Network CodingSun, HuaJafar, Syed2018-10-24We show that to characterize the capacity of storage overhead constrained private information retrieval (PIR) with only 2 messages, and 2 databases, non-Shannon information inequalities are necessary. As a by-product of this result, we construct the smallest instance, to our knowledge, of a network coding capacity problem that requires non-Shannon inequalities.Private Information RetrievalNon-Shannon Information InequalityStorage OverheadCapacityNetwork Codingapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/70d1v0jtpublicationoai:escholarship.org:ark:/13030/qt41h6v54h2018-01-23T17:18:21Zqt41h6v54hOptimality of Simple Layered Superposition Coding in the 3 User MISO BC with Finite Precision CSITGholami Davoodi, ArashJafar, Syed A2018-01-12We study the 3 user multiple input single output (MISO) broadcast channel (BC) with 3 antennas at the transmitter and 1 antenna at each receiver, from the generalized degrees of freedom (GDoF) perspective, under the assumption that the channel state information at the transmitter (CSIT) is limited to finite precision. In particular, our goal is to identify a parameter regime where a simple layered superposition (SLS) coding scheme achieves the entire GDoF region. With αij representing the channel strength parameter for the link from the jth antenna of the transmitter to the ith receiver, we prove that SLS is GDoF optimal without the need for time-sharing if max(αji,αij) ≤ αii and αki + αij ≤ αii + αkj for all i,j,k ∈ [3]. The GDoF region under this condition is a convex polyhedron.Generalized Degrees of FreedomBroadcast ChannelSuperposition CodingCSITapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/41h6v54hpublicationoai:escholarship.org:ark:/13030/qt5vc8f7h02017-10-31T17:44:10Zqt5vc8f7h0The Capacity of Private ComputationSun, HuaJafar, Syed A2017-10-31CapacityPrivate ComputationPrivate Information Retrievalapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/5vc8f7h0publicationoai:escholarship.org:ark:/13030/qt8cj3m85d2016-04-12T23:23:11Zqt8cj3m85dInterference Mitigation Using Asynchronous Transmission and Sampling DiversityGanji, MehdiJafarkhani, Hamid2016-04-12In this paper, we show that by investigating inherent time delays between different users in a multiuser scenario, we are able to cancel interference more efficiently. Time asynchrony provides another tool to cancel interference which results in preserving other resources like frequency, time and code. Therefore, we can save the invaluable resource of frequency band and also increase spectral efficiency. A sampling method is presented which results in independent noise samples and obviates the need for the complex process of noise whitening. By taking advantage of this sampling method and its unique structure, we implement maximum-likelihood sequence detection which outperforms synchronous maximum-likelihood detection. We also present successive interference cancellation with hard decision passing which gives rise to a novel forward-backward belief propagation method. Next, the performance of zero forcing detection is analyzed. Simulation results are also presented to verify our analysis.application/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/8cj3m85dpublicationoai:escholarship.org:ark:/13030/qt9cv1135b2016-02-12T16:36:55Zqt9cv1135bTransmitter Cooperation under Finite Precision CSIT: A GDoF PerspectiveDavoodi, Arash GJafar, Syed A2016-02-12The benefits of partial and full transmitter cooperation are evaluated for a two user interference channel under finite precision channel state information at the transmitters (CSIT), using the generalized degrees of freedom (GDoF) metric. Under finite precision CSIT, the benefits of interference alignment are completely lost, so that the $X$ channel obtained by partial transmitter cooperation does no better than the underlying interference channels. Full transmitter cooperation produces a vector broadcast channel (BC) which has a strict GDoF advantage over partial cooperation (X channel) and whose GDoF are fully achieved by interference enhancement. Generalized Degrees of FreedomInterference AlignmentAligned Image Setsapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/9cv1135bpublicationoai:escholarship.org:ark:/13030/qt1z2266042015-07-09T18:41:15Zqt1z226604Correction to “Multicast Networks with Variable-Length Limited Feedback”Liu, Xiaoyi (Leo)Koyuncu, ErdemJafarkhani, Hamid2015-07-01application/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/1z226604publicationoai:escholarship.org:ark:/13030/qt1c17p6zq2015-01-09T16:28:24Zqt1c17p6zqDegrees of Freedom of Rank-Deficient MIMO Interference NetworksKrishnamurthy, Sundar RajanRamakrishnan, AbineshJafar, Syed Ali2015-01-09We characterize the degrees of freedom (DoF) of MIMO interference networks with rank-deficient channel matrices. For the 2-user rank deficient MIMO interference channel, we provethe optimality of previously known achievable DoF in the symmetric case and generalize the re-sult to fully asymmetric settings. For the K-user rank deficient interference channel, we improvethe previously known achievable DoF and provide a tight outer bound to establish optimalityin symmetric settings. In particular, we show that for the K-user rank deficient interferencechannel, when all nodes have M antennas, all direct channels have rank D0, all cross chan-nels are of rank D, and the channels are otherwise generic, the optimal DoF value per user ismin(D0, M −min(M,(K−1)D)). For 2-user and 3-user rank deficient channels, achievable schemes2are for both constant and time-varying channels, while for K-user rank deficient channels, we present schemes for time-varying channels and note that the insights would act as stepping stones for constant channels. Notably for interference channels, the rank-deficiency of direct channels does not help and the rank-deficiency of cross-channels does not hurt. The main technical challenge is to account for the spatial dependencies introduced by rank deficiencies in the interference alignment schemes that typically rely on the independence of channel coefficients.Interference AlignmentDegrees of FreedomRank-Deficientapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/1c17p6zqarticleIEEE Transactions on Information Theoryvol 61, iss 1oai:escholarship.org:ark:/13030/qt0010s3qd2014-01-13T18:49:10Zqt0010s3qdTwo-User Multicast Networks with Variable-Length Limited Feedback Liu, XiaoyiKoyuncu, ErdemJafarkhani, Hamid2014-01-01application/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/0010s3qdpublicationoai:escholarship.org:ark:/13030/qt19d0q7jx2013-10-31T05:13:17Zqt19d0q7jxMultiscale Image Quality EstimationDemirtas, A. MuratReibman, Amy R.Jafarkhani, Hamid2013-10-30Multimedia communication is becoming pervasive because of the progress in wireless communications and multimedia coding. Estimating the quality of the visual content accurately is crucial in providing satisfactory service. State of the art visual quality assessment approaches are effective when the input image and the reference image have the same resolution. However, finding the quality of an image that has spatial resolution different than that of the reference image is still a challenging problem. To solve this problem, we develop a quality estimator (QE) which computes the quality of the input image without resampling the reference or the input images. In this work, we begin by identifying the potential weaknesses of previous approaches used to estimate the quality of experience. Next, we design a QE to estimate the quality of a distorted image with a lower resolution compared to the reference image. We also propose a subjective test environment to explore the success of the proposed algorithm in comparison with other QEs. When the input and test images have different resolutions, the subjective tests demonstrate that in most cases the proposed method works better than other approaches. In addition, the proposed algorithm also performs well when the reference image and the test image have the same resolution.image quality estimationhuman visual systempaired comparisonspatial resolutionsubjective testsapplication/pdfCC-BY-NC-NDeScholarship, University of Californiahttps://escholarship.org/uc/item/19d0q7jxpublicationoai:escholarship.org:ark:/13030/qt66b461fd2013-09-30T01:53:08Zqt66b461fdTopological Interference Management through Index CodingJafar, Syed A2013-01-16While much recent progress on interference networks has come about under the assumption of abundant channel state information at the transmitters (CSIT), a complementary perspective is sought in this work through the study of interference networks with no CSIT except a coarse knowledge of the topology of the network that only allows a distinction between weak and significant channels and no further knowledge of the channel coefficients' realizations. Modeled as a degrees-of-freedom (DoF) study of a partially connected interference network with no CSIT, the problem is found to have a counterpart in the capacity analysis of wired networks with arbitrary linear network coding at intermediate nodes, under the assumption that the sources are aware only of the end to end topology of the network. The network capacity (wired) and DoF (wireless) region, expressed in dimensionless units as multiples of the capacity (wired) and DoF (wireless) of a single link, are found to be bounded above by the capacity of an index coding problem where the antidotes graph is the complement of the interference graph of the original network and the bottleneck link capacity is normalized to unity. The problems are shown to be equivalent under linear solutions. An interference alignment perspective is then used to translate the existing index coding solutions into the wired network capacity and wireless network DoF solutions, as well as to find new and unified solutions to different classes of all three problems. For networks with $K$ messages, a study of the extremes -- when each message achieves half the cake, and when each message can achieve no more than $1/K$ of the cake, reveals the necessary and sufficient conditions for each, in terms of alignment graphs and demand graphs, respectively. Half the cake per message is achievable if and only if the alignment graph has no internal conflicts. No more than $1/K$ of the cake is achievable if and only if the network can be relaxed into a $K$-unicast setting with an acyclic demand graph. For half-rate-feasible networks, best case capacity (DoF) improvements over the best orthogonal (TDMA) and multicast (CDMA) solutions are explored for multiple groupcast and multiple unicast settings, and shown to be of polynomial order in the number of messages. For intermediate cases where neither half the cake, nor $1/K$ of the cake per message is capacity (DoF) optimal, the interference alignment perspective is used to characterize the symmetric capacity (DoF) of all cases where each alignment set either does not contain a cycle or does not contain a fork. A study of linear feasible rates shows duality properties that are used to extend the scope of previous results. For wireless networks, extensions to multiple antenna networks are made in symmetric settings where all nodes are equipped with the same number of antennas. The study of certain topologies of interest, motivated by cellular networks reveals interesting aligned frequency reuse patterns.interference alignmentindex codingapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/66b461fdpublicationoai:escholarship.org:ark:/13030/qt9qc3343h2013-01-10T23:27:22Zqt9qc3343hDegrees of Freedom of the Two-Way Relay MIMO Interference ChannelWang, ChenweiJafar, Syed A2013-01-10We investigate the symmetric degrees of freedom (DoF) of the K-pair (2K users) two-way relay Multiple-Input Multiple-Output (MIMO) Gaussian interference channel for K = 2,3 where each user is equipped with M antennas and the relay node is equipped with N antennas. The two users of each pair communicate with each other via the help of the relay only. Expressing the DoF characterization as a function of the ratio γ = M/N, we find that the DoF value per user is piecewise linear depending on M and N alternately. As we will show in this paper, while the DoF achievability only needs linear beamforming transmission and zero-forcing reception, inter-pair signal subspace alignment is essential at the relay node as well as the users. In addition, the DoF converse is first developed based on the linear dimension counting approach, which can be further translated to the information theoretic statement.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/9qc3343hpublicationoai:escholarship.org:ark:/13030/qt2nf4s7hf2012-11-08T22:30:19Zqt2nf4s7hfDegrees of Freedom of the Interference Channel with a Cognitive HelperWang, ChenweiSezgin, Aydin2012-11-08Abstract—In this letter, we characterize the degrees of freedom (DoF) of theK≥3user Gaussian interference network with a cognitive helper where each node is equipped with only one antenna. Specifically, each user sends one independent message to its corresponding receiver through its own antenna and via the help of the cognitive helper. For this network, we show that the sum DoF value is outer bounded by(K + 1)/2whenKis odd andK2/(2(K + 1))whenKis even, respectively. The new DoF outer bounds are derived based on the fact that collaboration among users does not decrease the capacity region and increasing the number of users does not increase the capacity per user. In addition, we provide a new achievable scheme to achieve a total of(K + 1)/2DoF for anyK≥3. Thus, the exact DoF value of the network is characterized with the total DoF given as(K + 1)/2, wheneverKis odd. The new achievable scheme is based on interference neutralization and asymptotic interference alignment.Degrees of FreedomCognitive communicationInterference Channelapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/2nf4s7hfpublicationoai:escholarship.org:ark:/13030/qt3gc1z8kn2012-11-08T18:05:20Zqt3gc1z8knDegrees of Freedom of 2-user and 3-user Rank-Deficient MIMO Interference ChannelsKrishnamurthy, Sundar RajanJafar, Syed Ali2012-04-02We study the degrees of freedom (DoF) of 2-user and 3-user multiple input multiple output (MIMO) interference channels with rank deficient channel matrices. Only achievable DoF results and trivial outer bounds were previously available for these problems, restricted to symmetric settings. For the 2- user rank deficient MIMO interference channel we prove the optimality of previously known achievable DoF in the symmetric case and generalize the result to fully asymmetric settings. For the 3-user rank deficient MIMO interference channel, we improve the achievable DoF and provide a tight outer bound to establish optimality. Linear precoding based achievable schemes are found to be DoF optimal in both cases.Degrees of FreedomInterference AlignmentMIMOapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/3gc1z8knpublicationoai:escholarship.org:ark:/13030/qt72b5q2sf2012-04-22T19:25:48Zqt72b5q2sfOn Optimal Ergodic Interference AlignmentGeng, ChunhuaJafar, Syed Ali2012-04-02The original ergodic interference alignment scheme proposed by Nazer et al. requires symmetric channel phase dis- tribution. In this paper, we investigate a new ergodic interference alignment scheme which can achieve one half interference-free degree of freedom (DoF) for arbitrary phase distribution. Even for symmetric phase distributions, the new scheme achieves a better high SNR offset than the original ergodic interference alignment scheme, and depending upon the magnitude distribu- tions it is shown that the SNR offset improvement with the new scheme over the original scheme can be arbitrarily large. The SNR offset optimal ergodic alignment scheme is based on results in majorization theory.Ergodic interference alignmenthigh SNR offsetapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/72b5q2sfpublicationoai:escholarship.org:ark:/13030/qt02w920102011-10-26T16:36:34Zqt02w92010Feedback improves the generalized degrees of freedom of the strong interference channelCadambe, Viveck RJafar, Syed A2008-04-01We provide inner and outer bounds on the generalized degrees of freedom of the two user symmetric interference channel. The bounds are tight in the moderately weak and strong interference regimes. Feedback is shown to provide unbounded improvements to the generalized degrees of freedom of the two user interference channel in the very strong interference regime. It is also shown that feedback does not improve the GDOF of the channel if the interference is moderately weak or moderately strong. We also finally present bounds for the generalized degrees of freedom of the symmetric MIMO interference channel with feedback.(This report is the exact paper submitted to IEEE GLOBECOM for review, April 1, 2008 on EDAS. Uploaded, for the record, as CPCC report in Oct. 2011.)application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/02w92010publicationoai:escholarship.org:ark:/13030/qt6t14c3612011-10-22T18:36:50Zqt6t14c361On Optimality of Linear Interference Alignment for the Three-User MIMO Interference Channel with Constant Channel CoefficientsWang, ChenweiGou, TiangaoJafar, Syed Ali2011-10-22We investigate the optimality of linear interferencealignment (allowing symbol extensions) for  3-user$M_T\times M_R$ MIMO interference channel where $M_T$ and $M_R$denote the number of antennas at each transmitter and each receiver,respectively, and the \emph{channel coefficients are held constant}. Recently, Wang et al. have conjectured that interference alignment based on linear beamforming using only proper Gaussian codebooks and possibly with symbol extensions, is sufficient to achieve the information theoretic DoF outer bound for all $M_T, M_R$ values exceptif $|M_T-M_R|=1$, $\min(M_T,M_R)\geq 2$. A partial proof of the conjecture is provided by Wang et al. for arbitrary $M_T, M_R$ values subject to a final numerical evaluation step that needs to be performed for each $M_T, M_R$ setting to complete the proof. The numerical evaluation step is also carried out explicitly by Wang et al. to settle the conjecture for all  $M_T, M_R$ values up to 10. For $|M_T-M_R|=1$, $\min(M_T,M_R)\geq 2$, Wang et al. show that interference alignment schemes based on linear beamforming with proper Gaussian signaling and symbol extensions are not sufficient to achieve the information-theoretic DoF outer bonds. In contrast, in this note we show, for all $M_T, M_R$ values up to 10, that interference alignment schemes based on linear beamforming over symbol extensions are enough to achieve the information theoretic DoF outer bounds for constant channels, if \emph{asymmetric complex signaling} is utilized. Based on this new insight, we conjecture that linear interference alignment is optimal for achieving the information theoretic DoF outer bounds for all $M_T, M_R$ values in the 3 user $M_T\times M_R$ MIMO interference channel with constant channel coefficients, except for the case $M_T=M_R=1$ where it is known that either time/frequency-varying channels or non-linear (e.g., rational alignment) schemes are required.Interference AlignmentBeamformingFeasibilityAsymmetric Complex Signalingapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/6t14c361publicationoai:escholarship.org:ark:/13030/qt548776qj2011-10-20T23:35:12Zqt548776qjExploiting Heterogeneous Channel Coherence Intervals for Blind Interference AlignmentJafar, Syed Ali2011-10-20We explore 5  network communication problems where the possibility of interference alignment, and consequently the total number of degrees of freedom (DoF) with channel uncertainty at the transmitters,  are unknown. These problems share the common property that in each case the best known outer bounds are essentially robust to channel uncertainty and represent the outcome with interference alignment, but the best inner bounds --- in some cases conjectured to be optimal ---  predict a total collapse of DoF, thus indicating the infeasibility of interference alignment under channel uncertainty at transmitters. Our main contribution is to introduce the idea of blind interference alignment. Specifically, we show that even with no knowledge of channel coefficient values at the transmitters, the knowledge of the channels' coherence structure can be exploited to achieve interference alignment. In each case, we show that under certain heterogeneous block fading models, i.e., when certain users experience smaller coherence time/bandwidth than others,  the transmitters are able to align interference without the knowledge of channel coefficient values.Interference AlignmentX ChannelInterference ChannelBroadcast Channelapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/548776qjpublicationoai:escholarship.org:ark:/13030/qt0112p0hf2011-07-04T03:57:55Zqt0112p0hfInterference Alignment and Degrees of Freedom of the Two-User X Channel with an Instantaneous RelayWang, ChenweiJafar, Syed A2011-06-11In this paper we investigate the sum degrees of freedom (DoF) of multiple unicasts in a wireless network. With 2 source nodes, 2 destination nodes, there are a total of 4 independent unicast sessions (messages), one from each source to each sink node (this setting is also known as an X network), and also there is a delay-free relay working in full-duplex mode helping the transmissions from the source to destination nodes. For such a channel setting, we prove that 5/3 DoF is achievable almost surely for time-varying/frequency-selective channels, based on the ideas of aligned interference neutralization, linear forwarding and interference alignment. Also, the achievable scheme can be easily translated to the rational alignment scheme for the network with constant-values channel coefficients. In addition, we provide an intuition for the 5/3 DoF result from the perspective of counting the number of linear equations and variables.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/0112p0hfpublicationoai:escholarship.org:ark:/13030/qt6z00q9qn2011-07-04T03:49:09Zqt6z00q9qnShort-term Performance Limits of MIMO Systems with Side Information at the TransmitterLi, LiangbinJafarkhani, Hamid2011-06-03The fundamental performance limits of space-time block code (STBC) designs when perfect channel information is available at the transmitter (CSIT) are studied in this report. With CSIT, the transmitter can perform various techniques such as rate adaption, power allocation, or beamforming. Previously, the exploration of these fundamental results assumed long-term constraints, for example, channel codes can have infinite decoding delay, and power or rate is normalized over infinite channel-uses. With long-term constraints, the transmitter can operate at the rate lower than the instantaneous mutual information and error-free transmission can be supported. In this report, we focus on the performance limits of short-term behavior for STBC systems. We assume that the system has block power constraint, block rate constraint, and finite decoding delay. With these constraints, although the transmitter can perform rate adaption, power control, or beamforming, we show that decoding-error is unavoidable. In the high SNR regime, the diversity gain is upperbounded by the product of the number of transmit antennas, receive antennas, and independent fading block channels that messages spread over. In other words, fading cannot be completely combatted with short-term constraints. The proof is based on a sphere-packing argument.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/6z00q9qnpublicationoai:escholarship.org:ark:/13030/qt7j98g3qj2011-07-04T02:51:33Zqt7j98g3qjAligned Interference Neutralization and the Degrees of Freedom of the 2x2x2 Interference ChannelGou, Tiangao2011-02-26We show that the 2 × 2 × 2 interference network, i.e., the multihop interference network formed by concatenation of two 2-user interference channels achieves the min-cut outer bound value of 2 DoF, for almost all values of channel coefficients, for both time-varying or fixed channel coefficients. The key to this result is a new idea, called aligned interference neutralization, that provides a way to align interference terms over each hop in a manner that allows them to be cancelled over the air at the last hop.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/7j98g3qjpublicationoai:escholarship.org:ark:/13030/qt4b2338qk2011-07-04T00:52:15Zqt4b2338qkThe Ergodic Capacity of Interference NetworksJafar, Syed A2010-12-01We identify the role of equal strength interference links as bottlenecks on the ergodic sum capacity of a K user interference network with the fading process restricted primarily to inde- pendent and uniform phase variations while the channel magnitudes are held fixed across time. It is shown that even though there are K(K − 1) cross-links, only about K/2 disjoint and equal strength interference links suffice to determine the capacity of the network regardless of the strengths of the rest of the cross channels. This scenario is called a minimal bottleneck state. It is shown that ergodic interference alignment is capacity optimal for a network in a minimal bottleneck state. The results are applied to large networks. It is shown that large networks are close to bottleneck states with a high probability, so that ergodic interference alignment is close to optimal for large networks. Limitations of the notion of bottleneck states are also high- lighted for channels where both the phase and the magnitudes are allowed to vary with time. It is shown through an example that for these channels, joint coding across different bottleneck states makes it possible to circumvent the capacity bottlenecks.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/4b2338qkpublicationoai:escholarship.org:ark:/13030/qt475040fd2011-07-04T00:50:00Zqt475040fdDiversity Results for DSTC-ICRec and DSTC Joint-user ML decodingLi, LiangbinJing, YindiJafarkhani, Hamid2010-01-01In this technical report, we provide diversity analysis for two transmission schemes in (J;Ja;Ra;N) multi-access relay networks (MARNs), where J users, each equipped with Ja antennas, communicate to one N-antenna receiver through Ra single-antenna relays. Both transmission schemes allow all users' symbols to be transmitted concurrently through the source-relay and relay-receiver links. Therefore, both schemes have the potential of high transmission rate in multi-user relay networks. In the first scheme, called DSTC joint-user ML decoding, the relays perform distributed space-time coding (DSTC) to improve the reliability of the system, and the receiver jointly decodes all users' symbols using the ML decoding. Through rigorous analysis, this scheme achieves a symbol rate of 1/2 symbols/user/channel use, in conjunction to a diversity gain of Ra*min(Ja;N), which is the maximum diversity achievable in this network. But the decoding complexity of this scheme is exponential in the number of users. To reduce the decoding complexity, we consider a second scheme, DSTC-ICRec, in which the relay operations are the same, but the receiver first conducts interference cancellation (IC) to decouple multi-user signals, then decodes each user's symbols independently. We show analytically that in (2; 1; 2; N) and (2; 2;2;N) MARNs, DSTC-ICRec achieves a diversity of 1 and min(2;N-1), respectively, at a symbol rate of 1/2 symbols/user/channel use. Since the maximum achievable diversity gains in these two networks are 2 and 2*min(2;N), respectively, DSTC-ICRec has a lower diversity gain, compared to DSTC joint-user ML decoding, but its decoding complexity is much lower due to the IC.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/475040fdpublicationoai:escholarship.org:ark:/13030/qt926746j92011-07-03T23:56:35Zqt926746j9The Necessity of Relay SelectionKoyuncu, ErdemJafarkhani, Hamid2010-11-26We determine necessary conditions on the structure of symbol error rate (SER) optimal quantizers for limited feedback beamforming in wireless networks with one transmitter-receiver pair and R parallel amplify-and-forward relays. We call a quantizer codebook "small" if its cardinality is less than R, and "large" otherwise. A "d-codebook" depends on the power constraints and can be optimized accordingly, while an "i-codebook" remains fixed. It was previously shown that any i-codebook that contains the single-relay selection (SRS) codebook achieves the full-diversity order, R. We prove the following: Every full-diversity i-codebook contains the SRS codebook, and thus is necessarily large. In general, as the power constraints grow to infinity, the limit of an optimal large d-codebook contains an SRS codebook, provided that it exists. For small codebooks, the maximal diversity is equal to the codebook cardinality. Every diversity-optimal small i-codebook is an orthogonal multiple-relay selection (OMRS) codebook. Moreover, the limit of an optimal small d-codebook is an OMRS codebook. We observe that SRS is nothing but a special case of OMRS for codebooks with cardinality equal to R. As a result, we call OMRS as "the universal necessary condition" for codebook optimality. Finally, we confirm our analytical findings through simulations.Relay SelectionFeedbackBeamformingDiversityapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/926746j9publicationoai:escholarship.org:ark:/13030/qt75s218262011-07-03T23:56:30Zqt75s21826Distributed Beamforming in Wireless Multiuser Relay-Interference Networks with Quantized FeedbackKoyuncu, ErdemJafarkhani, Hamid2010-11-26We study quantized beamforming in wireless amplify-and-forward relay-interference networks with any number of transmitters, relays, and receivers. We design the quantizer of the channel state information to minimize the probability that at least one receiver incorrectly decodes its desired symbol(s). Correspondingly, we introduce a generalized diversity measure that encapsulates the conventional one as the first-order diversity. Additionally, it incorporates the second-order diversity, which is concerned with the transmitter power dependent logarithmic terms that appear in the error rate expression. First, we show that, regardless of the quantizer and the amount of feedback that is used, the relay-interference network suffers a second-order diversity loss compared to interference-free networks. Then, two different quantization schemes are studied: First, using a global quantizer, we show that a simple relay selection scheme can achieve maximal diversity. Then, using the localization method, we construct both fixed-length and variable-length local (distributed) quantizers (fLQs and vLQs). Our fLQs achieve maximal first-order diversity, whereas our vLQs achieve maximal diversity. Moreover, we show that all the promised diversity and array gains can be obtained with arbitrarily low feedback rates when the transmitter powers are sufficiently large. Finally, we confirm our analytical findings through simulations.BeamformingFeedbackRelay NetworkInterference Networkapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/75s21826publicationoai:escholarship.org:ark:/13030/qt7cg6q1gw2011-07-03T23:41:04Zqt7cg6q1gwDegrees of Freedom of the Cellular System with Relays and Partial CSITGou, TiangaoJafar, Syed A2010-11-12We consider a two-tier cellular system where there is a macrocell base station equipped with multiple antennas transmitting to multiple microcells, each equipped with a microcell base station (relay) equipped with one antenna. Each microcell has one user. We explore the degrees of freedom (DoF) of several configurations of this network with the assumption that no channel state information (CSI) of the users is available at the macrocell base station and different assumptions on the CSI between other transmitter-receiver pairs. With very limited CSI and the relays, the achievable DoF can be greatly increased compared to the case when there is no relays.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/7cg6q1gwpublicationoai:escholarship.org:ark:/13030/qt6cp9n6s12011-07-03T23:37:01Zqt6cp9n6s1Multiple-Antenna Interference Cancellation and Detection for Two Users Using Quantized FeedbackLi, FengJafarkhani, Hamid2010-09-27When two users transmit signals to a common receiver, one can design precoders to cancel the interference for each user, if each user knows all the channel information perfectly. Also the diversity for each user is full. However, in practice, perfect channel information is not available. In this paper, we design precoders for two users with two transmit antennas and one receiver with two receive antennas using quantized feedback. We propose to construct codebook using Grassmannian line packing. By choosing precoders from the codebook properly, our proposed scheme can cancel the interference for each user. Also we analytically prove that our system can achieve full diversity for each user. Then we extend our scheme to any number of transmit and receive antennas. Simulation results confirm our analytical proof and show that our scheme can serve as a bridge between a system with no feedback and a system with perfect feedback.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/6cp9n6s1publicationoai:escholarship.org:ark:/13030/qt8hs1d0tt2011-07-03T23:23:32Zqt8hs1d0ttInterference Cancellation at the Relay for Multi-User Wireless Cooperative NetworksLi, LiangbinJing, YindiJafarkhani, Hamid2010-09-21We study multi-user transmission and detection schemes for a multi-access relay network (MARN) with linear constraints at all nodes. In a (J, Ja, Ra, M) MARN, J sources, each equipped with Ja antennas, communicate to one M-antenna destination through one Ra-antenna relay. A new protocol called IC-Relay-TDMA is proposed which takes two phases. During the first phase, symbols of different sources are transmitted concurrently to the relay. At the relay, interference cancellation (IC) techniques, previously proposed for systems with direct transmission, are applied to decouple the information of different sources without decoding. During the second phase, symbols of different sources are forwarded to the destination in a time division multi-access (TDMA) fashion. At the destination, the maximum-likelihood (ML) decoding is performed source-by-source. The protocol of IC-Relay-TDMA requires the number of relay antennas no less than the number of sources, i.e., Ra >= J. Through outage analysis, the achievable diversity gain of the proposed scheme is shown to be min{Ja(Ra â^È^Ò J + 1), RaM}. When M â^ɤ Ja (1 â^È^Ò (Jâ^È^Ò1) / Ra), the proposed scheme achieves the maximum interference-free (int-free) diversity gain RaM. Since concurrent transmission is allowed during the first phase, compared to full TDMA transmission, the proposed scheme achieves the same diversity, but with a higher symbol rate.Multi-access relay networkdistributed space-time codinginterference cancellationorthogonal designsquasi-orthogonal designscooperative diversityapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/8hs1d0ttpublicationoai:escholarship.org:ark:/13030/qt9511q6pf2011-07-03T23:23:27Zqt9511q6pf"Using Instantaneous Normalized Receive SNR for Diversity Gain CalculationLi, LiangbinJing, YindiJafarkhani, Hamid2010-09-21We propose a technique to calculate the diversity gain for a vector channel when the noises have any nonsingular distribution and the fading coefficients have arbitrary distribution. This technique uses the serial expansion of the outage probability of the instantaneous normalized receive SNR to obtain the diversity gain. The approach is simpler than using the pairwise symbol error rate to analyze the diversity gain.application/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/9511q6pfpublicationoai:escholarship.org:ark:/13030/qt40m956wj2011-07-03T23:18:40Zqt40m956wjRetrospective Interference AlignmentMaleki, HamedJafar, Syed AShamai, Shlomo2010-09-17We explore similarities and differences in recent works on blind interference alignment under different models such as staggered block fading model and the delayed CSIT model. In particular we explore the possibility of achieving interference alignment with delayed CSIT when the transmitters are distributed. Our main contribution is an interference alignment scheme, called retrospective interference alignment in this work, that is specialized to settings with distributed transmitters. With this scheme we show that the 2 user X channel with only delayed channel state information at the transmitters can achieve 8/7 DoF, while the interference channel with 3 users is able to achieve 9/8 DoF. We also consider another setting where delayed channel output feedback is available to transmitters. In this setting the X channel and the 3 user interference channel are shown to achieve 4/3 and 6/5 DoF, respectively.Interference AlignmentDegrees of FreedomX channelCSITapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/40m956wjpublicationoai:escholarship.org:ark:/13030/qt28s196nf2011-07-02T07:37:16Zqt28s196nfVOFDM Broadband Wireless Transmission and Its AdvantagesAyanoglu, EnderJones, VKRaleigh, Gregory G.Gardner, JamesGerlach, DerekToussi, Karim2001-02-06In this paper we describe a coding, modulation, and spatial processing technique for fixed broadband wireless Internet access applications and provide examples of its performance. This technique is built on Orthogonal Frequency Division Multiplexing (OFDM) and is known as Vector OFDM (VOFDM). We compare VOFDM with conventional Single Carrier Modulation (SCM), and show that it provides substantial performance improvements over SCM. 1. It divides the channel into narrowband, flat fading, subchannels and thus it is more resistant to frequency selective fading as compared to single carrier systems. 2. By using FFT techniques, it is computationally efficient. 3. It can be combined with coding and interleaving to recover symbols lost due to the frequency selectivity ofapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/28s196nfpublicationoai:escholarship.org:ark:/13030/qt68w383002011-07-02T07:37:06Zqt68w38300B-ISDN (Broadband Integrated Services Digital Network)Ayanoglu, EnderAkar, Nail2002-05-25The subject of B-ISDN came into being in the late 1980s, together with the concept of Asynchronous Transfer Mode (ATM). ATM is closely tied to high-speed packet switching by means of specialized switches implemented in hardware. Due to its high speed and packet structure, ATM technology was considered attractive to unify voice, data, and video services. A unification of these services over the telephone infrastructure was attempted earlier by a standards offering known as Integrated Services Digital Network (ISDN). Consequently, this new service unification was termed Broadband ISDN (B-ISDN). Although due to its origins, B-ISDN is sometimes closely tied to ATM technology, the term independently represents the vision of packet-based high-speed integration of voice, data, and video services. It is important that in this process, guarantees to satisfy different Quality-of-Service (QoS) needs (in terms of delay, loss, etc) required by voice, data, and video services are provided. In this vision, what is important is the unification, or integration of services; and the underlying technology is of secondary importance. As of the early 2000s, the technology to be employed in realizing this vision seems to have shifted from its origins of ATM. In this article, our emphasis is on B-ISDN as a service integration vision. Nevertheless, we will describe its original emphasis as the service offering of ATM as well as the path the industry seems to be taking in implementing this vision.ISDNB-ISDNATMIPQoSITU-Tapplication/pdfpubliceScholarship, University of Californiahttps://escholarship.org/uc/item/68w38300publication