On the capacity of cloud radio access networks with oblivious relaying

Uplink and downlink cloud radio access networks are modeled as two-hop K-user L-relay networks, whereby small base-stations act as relays for end-to-end communications and are connected to a central processor via orthogonal fronthaul links of finite capacities. Simplified versions of network compressforward (or noisy network coding) and distributed decode-forward are presented to establish inner bounds on the capacity region for uplink and downlink communications, that match the respective cutset bounds to within a finite gap independent of the channel gains and signal to noise ratios. These approximate capacity regions are then compared with the capacity regions for networks with no capacity limit on the fronthaul. Although it takes infinite fronthaul link capacities to achieve these "fronthaul-unlimited" capacity regions exactly, these capacity regions can be approached approximately with finite-capacity fronthaul. The total fronthaul link capacities required to approach the fronthaul-unlimited sum-rates (for uplink and downlink) are characterized. Based on these results, the capacity scaling law in the large network size limit is established under certain uplink and downlink network models, both theoretically and via simulations. A. Uplink C-RANSeveral coding schemes have been proposed in the literature for the uplink C-RAN with K users (senders) and L relays. Zhou and Yu [5] applied the network compress-forward relaying scheme [7] to this model and showed, by optimizing over quantizers, that under some symmetry assumptions, this scheme achieves the optimal sum-rate within L/2 bits per real dimension uniformly over all K and all channel parameters. Sanderovich, Someskh, Poor, and Shamai [8] used the same scheme and analyzed the large-user asymptotics (i.e., the scaling law) of symmetric achievable rates when all fronthaul links have equal capacities. Zhou, Xu, Yu, and Chen [6] subsequently showed that under a sum-capacity constraint on the fronthaul links, the coding scheme in [5] and [8] can be simplified through successive cancellation decoding, generalizing an earlier result for the single-sender multiple relay network [9]. Aguerri and Zaidi [10] proposed a hybrid coding scheme of network compress-forward and compute-forward [11], and demonstrated that it outperforms the better of the two in general. Aguerri, Zaidi, Caire, and Shamai [12] specialized the noisy network coding scheme [13] to the uplink C-RAN, the achievable rate region of which coincides with that of network compress-forward [5], [8]. The most general outer bound on the capacity region of the uplink C-RAN can be obtained by specializing the cutset bound [14]; see, for example, [5] and the references therein, as well as Proposition 2 in this paper. The cutset bound has been further tightened under additional assumptions. Aguerri, Zaidi, Caire, and Shamai [12] studied the uplink C-RAN in which the relays are oblivious of the codebooks of the senders, and demonstrated that network compress-forward (or noisy network coding) achieves the capacit...

show abstract

“…The same inner bound can also be achieved by specializing [12] the noisy network coding scheme [13].…”

Section: Fig 2: Uplink Network Modelmentioning

confidence: 88%

“…, y L ) : y l ≤ỹ l , l ∈ [L]} with corner pointỹ. Moreover, this cuboid includes the point y * , sinceỹ l ≥ y * l for each l by (12). Thus, the point y * lies in the intersection P(φ) ∩ P(ψ) and therefore, network compress-forward, with this choice of C 1 , .…”

Section: Comparisons With Fronthaul-unlimited Uplinkmentioning

confidence: 97%

Capacity Scaling for Cloud Radio Access Networks with Limited Orthogonal Fronthaul

Ganguly

Kim

2019

2019 IEEE International Symposium on Information Theory (ISIT)

View full text Add to dashboard Cite

show abstract

“…, τ p } and data signal y ul l to the CPU, which takes care of channel estimation and data detection in a centralized fashion. In other words, the APs act as relays that forward all signals to the CPU [46]. In each coherence block, AP l needs to send τ p N complex scalars for the pilot signals and τ u N complex scalars for the received signals.…”

Section: B Uplink Data Transmissionmentioning

confidence: 99%

Scalable Cell-Free Massive MIMO Systems

Björnson

Sanguinetti

2020

IEEE Trans. Commun.

626

583

View full text Add to dashboard Cite

Imagine a coverage area with many wireless access points that cooperate to jointly serve the users, instead of creating autonomous cells. Such a cell-free network operation can potentially resolve many of the interference issues that appear in current cellular networks. This ambition was previously called Network MIMO (multiple-input multiple-output) and has recently reappeared under the name Cell-Free Massive MIMO. The main challenge is to achieve the benefits of cell-free operation in a practically feasible way, with computational complexity and fronthaul requirements that are scalable to large networks with many users. We propose a new framework for scalable Cell-Free Massive MIMO systems by exploiting the dynamic cooperation cluster concept from the Network MIMO literature. We provide algorithms for initial access, pilot assignment, cluster formation, precoding, and combining that are proved to be scalable. Interestingly, the proposed scalable precoding and combining outperform conventional maximum ratio processing and also performs closely to the best unscalable alternatives. Index TermsCell-Free Massive MIMO, scalable implementation, centralized and distributed algorithms, dynamic cooperation clustering, user-centric networking. 2 impractical, assumptions that lead to immense fronthaul signaling for CSI and data sharing, respectively, as well as huge computational complexity. Fortunately, [9] proved that Network MIMO can operate without CSI sharing, by sacrificing the ability for the APs to jointly cancel interference. Moreover, to limit data sharing and computational complexity, each UE can be served only by an AP subset [10]. Initially, a network-centric approach was taken by dividing the APs into non-overlapping (disjoint) cooperation clusters in which the APs are sharing data (and potentially CSI) to serve only the UEs residing in the joint coverage area [11]-[13]. This approach was considered in 4G but provides small practical gains [14]. One key reason is that many UEs will be located at the edges of the clusters and, thus, will observe substantial intercluster interference from the neighboring clusters [15].The alternative is to take a user-centric approach where each UE is served by the AP subset providing the best channel conditions. Since these subsets are generally different for every UE, it is not possible to divide the network into non-overlapping cooperation clusters. Instead, each AP needs to cooperate with different APs when serving different UEs, over the same time and frequency resource [16]-[18]. 1 A general user-centric cooperation framework was proposed in [17] under the name dynamic cooperation clustering (DCC) and was further described and analyzed in the textbook [10]. The word dynamic refers to the adaptation to time-variant characteristics such as channel properties and UE locations (to name a few). The practical feasibility of DCCs was experimentally verified by the pCell technology [21], but the combination of Network MIMO and DCC didn't gain much interest at the time it was prop...

show abstract

“…with the 0 in its principal diagonal denoting the n 0 ×n 0 -all zero matrix. In particular, for the setting of testing against independence, i.e., Y 0 = ∅ and the decoder's task reduced to guessing whether Y 1 and X are independent or not, the optimal trade-off expressed by (19) reduces to the set of (R 1 , E) pairs that satisfy, for some…”

Section: Memoryless Vector Gaussian Casementioning

confidence: 99%

“…Observe that (19) is the counter-part, to the vector Gaussian setting, of the result of [12,Theorem 3] which provides a single-letter formula for the Type II error exponent for the one-encoder DM testing against conditional independence problem. Similarly, (21) is the solution of the vector Gaussian version of the one-encoder DM testing against independence problem which is studied, and solved, by Ahlswede and Csiszar in [2, Theorem 2].…”

Section: Memoryless Vector Gaussian Casementioning

confidence: 99%