Hybrid architectures are generally composed of a cyber cloud with additional support of edge caching. By utilizing the benefits associated with cloud computing and caching, powerful enhanced interference management techniques can be readily utilized - that among others - also results in low-latency transmission. In this paper, we study the impact of cloud and edge processing on the latency for a heterogenous network (HetNet) consisting of two users and two transmitters. We define an information-theoretic metric, the delivery time per bit (DTB), that captures the delivery latency. We establish bounds on the DTB as a function of cache size, backhaul capacity and wireless channel parameters. We show the optimality on the DTB for various channel regimes.

The optimality of the simple scheme of treating interference as noise (TIN) is studied in this paper for the 2 × M linear deterministic (LD) X-channel. A new capacity upper bound is derived. In the considered scheme (denoted as 2-IC-TIN), the setup is reduced to a 2-user interference channel while the receivers employ TIN. It is shown that as long as 2-IC-TIN is optimal in a M × 2 X-channel, it is also capacity-optimal in the 2 × M X-channel which is generated by changing the role of transmitters and receivers. The result of this paper expands the capacity optimal regime of TIN for the 2 × M LD X-channel compared to the state of the art.

Stringent mobile usage characteristics force wireless networks to undergo a paradigm shift from conventional connection-centric to content-centric deployment. With respect to 5G, caching and heterogenous networks (HetNet) are key technologies that will facilitate the evolution of highly content-centric networks by facilitating unified quality of service in terms of low-latency communication. In this paper, we study the impact of transceiver caching on the latency for a HetNet consisting of a single user, a receiver and one cache-assisted transceiver. We define an information-theoretic metric, the delivery time per bit (DTB), that captures the delivery latency. We establish coinciding lower and upper bounds on the DTB as a function of cache size and wireless channel parameters; thus, enabling a complete characterization of the DTB optimality of the network under study. As a result, we identify cache beneficial and non-beneficial channel regimes.

This letter deals with the joint information and energy processing at the receiver of a point-to-point communication channel. In particular, the tradeoff between the achievable information rate and harvested energy for a multiple-antenna power splitting receiver is investigated. Here, the rate-energy region characterization is of particular interest, which is intrinsically a nonconvex problem. In this letter, an efficient algorithm is proposed for obtaining an approximate solution to the problem in polynomial time. This algorithm is mainly based on the Taylor approximation in conjunction with semidefinite relaxation, which is solved by interior-point methods. Moreover, we utilize the Gaussian randomization procedure to obtain a feasible solution for the original problem. It is shown that by proper receiver design the rate-energy region can be significantly enlarged compared to the state of the art, while at the same time the receiver hardware costs is reduced by utilizing less number of energy harvesting circuitry.

Treating interference as noise (TIN) as the most appropriate approach in dealing with interference and the conditions on its optimality has attracted the interest of researchers recently. However, our knowledge on necessary and sufficient conditions of TIN is restricted to a few setups with limited number of users. In this paper, we study the optimality of TIN in terms of the generalized degrees of freedom (GDoF) for a fundamental network, namely, the M × 2 X-channel. To this end, the achievable GDoF of TIN with power allocations at the transmitters is studied. It turns out that the transmit power allocation maximizing the achievable GDOF is given by ON-OFF signaling as long as the receivers use TIN. This leads to two variants of TIN, namely, P2P-TIN and 2-IC-TIN. While in the first variant the M × 2 X-channel is reduced to a point-topoint (P2P) channel, in the second variant, the setup is reduced to a two-user interference channel in which the receivers use TIN. The optimality of these two variants is studied separately. To this end, novel genie-aided upper bounds on the capacity of the X-channel are established. The conditions on the optimality of P2P-TIN can be summarized as follows. P2P-TIN is GDoF-optimal if there exists a dominant multiple access channel or a dominant broadcast channel embedded in the X channel. Furthermore, the necessary and sufficient conditions on the GDoF-optimality of 2-IC-TIN are presented. Interestingly, it turns out that operating the M × 2 X-channel in the 2-IC-TIN mode might be still GDOF optimal, although the conditions given by Geng et al. are violated. However, 2-IC-TIN is sub-optimal if there exists a single interferer which causes sufficiently strong interference at both receivers. The comparison of the results with the state of the art shows that the GDOF optimality of TIN is expanded significantly.

Decreasing communication latency is one of the main challenges for many applications in future communication systems. As a solution, we study in this work the benefits of using some cache-equipped helpers in a wireless communication scenario. More precisely, a wireless network with a transmitter, two cache-equipped helpers, and two receivers is studied from the latency point of view. While the transmitter has access to the whole data, the helpers store some data in their cache in order to support the transmitter in fulfilling the demands of the receivers. The main challenge is to develop a joint caching placement and the transmission strategy, which minimizes the latency of the communication. To this end, a general encoded caching is considered, in which each message is split into four types of sub-messages. The sub-messages of the first and second types are stored only in the cache of the first and second helper, respectively. The sub-messages of the third type are cached at both helpers, while the sub-messages of the fourth type are not cached at any helper. The size of the sub-messages is optimized in a way such that the delivery time of one bit for the worst-case demand is minimized. Moreover, a general lower bound on the delivery time is provided. Using this lower bound as a benchmark, the performance of the proposed caching policy is evaluated. It is shown that facilitating the cache-equipped helpers decreases the latency of broadcasting. Furthermore, it is shown that for the proposed caching policy, storing identical sub-messages in the cache of different helpers cannot be optimal if the cache size of the helpers is insufficiently large.

Despite the simplicity of the scheme of treating interference as noise (TIN), it was shown to be sum-capacity optimal in the Gaussian interference channel (IC) with very-weak (noisy) interference. In this paper, the two-user IC is altered by introducing an additional transmitter that wants to communicate with one of the receivers of the IC. The resulting network thus consists of a point-to-point channel interfering with a multiple access channel (MAC) and is denoted by PIMAC. The sum-capacity of the PIMAC is studied with main focus on the optimality of TIN. It turns out that TIN in its naive variant, where all transmitters are active and both receivers use TIN for decoding, is not the best choice for the PIMAC. In fact, a scheme that combines both time division multiple access and TIN (TDMA-TIN) strictly outperforms the naive-TIN scheme. Furthermore, it is shown that in some regimes, TDMA-TIN achieves the sum-capacity for the deterministic PIMAC and the sum-capacity within a constant gap for the Gaussian PIMAC. In addition, it is shown that, even for very-weak interference, there are some regimes where a combination of interference alignment with power control and TIN at the receiver side outperforms TDMA-TIN. As a consequence, on the one hand, TIN in a cellular uplink is approximately optimal in certain regimes. On the other hand, those regimes cannot be simply described by the strength of interference.

The impact of cooperation on interference management is investigated by studying an elemental wireless network, the so-called symmetric interference relay channel (IRC), from a generalized degrees of freedom (GDoF) perspective. This is motivated by the fact that the deployment of relays is considered as a remedy to overcome the bottleneck of current systems in terms of achievable rates. The focus of this paper is on the regime in which the interference link is weaker than the source-relay link in the IRC. Our approach toward studying the GDoF goes through the capacity analysis of the linear deterministic IRC (LD-IRC). New upper bounds on the sum capacity of the LD-IRC based on genie-aided approaches are established. These upper bounds together with some existing upper bounds are achieved by using four novel transmission schemes. Extending the upper bounds and the transmission schemes to the Gaussian case, the GDoF of the Gaussian IRC is characterized for the aforementioned regime. This completes the GDoF results available in the literature for the symmetric GDoF. It turns out that even if the incoming and outgoing links of the relay are both weaker than the desired channel, involving a relay can increase the GDoF. Interestingly, utilizing the relay in this case can increase the slope of the GDoF from -2 [in the interference channel (IC)] to -1 or 0. This shrinks the regime where ignoring the interference by treating it as noise is optimal. Furthermore, the analysis shows that if the relay ingoing and outgoing links are sufficiently strong, the relay is able to neutralize the interference completely. In this case, the bottleneck of the transmission will be the interference links, and hence, the GDoF increases if the interference link gets stronger. It is shown that in the strong interference regime, in contrast to the IC, the GDoF can be a monotonically decreasing function of the interference level.

The sum-capacity of a three user interference wired network for time-varying channels is considered. Due to the channel variations, it is assumed that the transmitters are only able to track the connectivity between the individual nodes, thus only the (alternating) state of the network is known. By considering a special subset of all possible states, we show that state splitting combined with joint encoding over the alternating states is required to achieve the sum-capacity. Regarding upper bounds, we use a genie aided approach to show the optimality of this scheme. This highlights that more involved transmit strategies are required for characterizing the degrees of freedom even if the transmitters have heavily restricted channel state information.

Ad-hoc networks that support multi-hopping can provide a range of solutions for range-limited sensor and machine-to-machine networks. However such networks have their own set of implementation and performance issues. Clustering helps in organizing the networks and providing a backbone for routing data packets over multi-hop links. However multi-hopping itself can cause a loss in performance compared to single-hop systems. This is because, as shown in the paper, the ergodic capacity of links in such networks is a decreasing function of the number of hops. Also error-propagation over multiple hops can result in a much reduced performance. Hence it is beneficial to reduce the number of hops in multi-hop networks. One way to do this is to employ cooperative MIMO (Multiple Input Multiple Output) techniques to increase the range of these devices. This will result in reducing the number of hops. However incorporating cooperative MIMO in an ad-hoc network without central control and with self-organised routing can be challenging. In this paper, we present a novel joint clustering and routing mechanism to simplify and define the procedure for nodes to employ cooperative MIMO to gain an increase in performance. We introduce the concept of 'cooperative-MIMO-linked neighboring clusters' and describe how that can be used in conjunction with our specialized routing mechanism AOCMR (Ad-hoc On-Demand Cooperative MIMO Routing), to enable nodes to efficiently communicate in large scale networks in a decentralized manner.

The concept of Ad-hoc mobile networks that support multi-hopping has been around for some time now. However, making sure that such decentralized networks are completely self-organizing is not a trivial task and raises various issues relating to network management, routing, interference, etc. We have built a software SONIR (Self-Organizing Network with Intelligent Relaying), in MATLAB, which implements an end-to-end multi-hop, virtual MIMO system, capable of dealing with mobility of nodes in a Rayleigh fading environment. Different methods for clustering, mobility managment, routing, virtual MIMO, etc. have been implemented. These methods work on different OSI layers. Our main goal is to be able to visualize such a system as a whole in order to see the end-to-end performance as well as solve the possible issues that arise with it.