-
Device-to-Device Coded Caching with Distinct Cache Sizes
Authors:
Abdelrahman M. Ibrahim,
Ahmed A. Zewail,
Aylin Yener
Abstract:
This paper considers a cache-aided device-to-device (D2D) system where the users are equipped with cache memories of different size. During low traffic hours, a server places content in the users' cache memories, knowing that the files requested by the users during peak traffic hours will have to be delivered by D2D transmissions only. The worst-case D2D delivery load is minimized by jointly desig…
▽ More
This paper considers a cache-aided device-to-device (D2D) system where the users are equipped with cache memories of different size. During low traffic hours, a server places content in the users' cache memories, knowing that the files requested by the users during peak traffic hours will have to be delivered by D2D transmissions only. The worst-case D2D delivery load is minimized by jointly designing the uncoded cache placement and linear coded D2D delivery. Next, a novel lower bound on the D2D delivery load with uncoded placement is proposed and used in explicitly characterizing the minimum D2D delivery load (MD2DDL) with uncoded placement for several cases of interest. In particular, having characterized the MD2DDL for equal cache sizes, it is shown that the same delivery load can be achieved in the network with users of unequal cache sizes, provided that the smallest cache size is greater than a certain threshold. The MD2DDL is also characterized in the small cache size regime, the large cache size regime, and the three-user case. Comparisons of the server-based delivery load with the D2D delivery load are provided. Finally, connections and mathematical parallels between cache-aided D2D systems and coded distributed computing (CDC) systems are discussed.
△ Less
Submitted 19 March, 2019;
originally announced March 2019.
-
Benefits of Coded Placement for Networks with Heterogeneous Cache Sizes
Authors:
Abdelrahman M. Ibrahim,
Ahmed A. Zewail,
Aylin Yener
Abstract:
In this work, we study coded placement in caching systems where the users have unequal cache sizes and demonstrate its performance advantage. In particular, we propose a caching scheme with coded placement for three-user systems that outperforms the best caching scheme with uncoded placement. In our proposed scheme, users cache both uncoded and coded pieces of the files, and the coded pieces at th…
▽ More
In this work, we study coded placement in caching systems where the users have unequal cache sizes and demonstrate its performance advantage. In particular, we propose a caching scheme with coded placement for three-user systems that outperforms the best caching scheme with uncoded placement. In our proposed scheme, users cache both uncoded and coded pieces of the files, and the coded pieces at the users with large memories are decoded using the unicast/multicast signals intended to serve users with smaller memories. Furthermore, we extend the proposed scheme to larger systems and show the reduction in delivery load with coded placement compared to uncoded placement.
△ Less
Submitted 9 November, 2018;
originally announced November 2018.
-
Coded Caching for Heterogeneous Systems: An Optimization Perspective
Authors:
Abdelrahman M. Ibrahim,
Ahmed A. Zewail,
Aylin Yener
Abstract:
In cache-aided networks, the server populates the cache memories at the users during low-traffic periods, in order to reduce the delivery load during peak-traffic hours. In turn, there exists a fundamental trade-off between the delivery load on the server and the cache sizes at the users. In this paper, we study this trade-off in a multicast network where the server is connected to users with uneq…
▽ More
In cache-aided networks, the server populates the cache memories at the users during low-traffic periods, in order to reduce the delivery load during peak-traffic hours. In turn, there exists a fundamental trade-off between the delivery load on the server and the cache sizes at the users. In this paper, we study this trade-off in a multicast network where the server is connected to users with unequal cache sizes and the number of users is less than or equal to the number of library files. We propose centralized uncoded placement and linear delivery schemes which are optimized by solving a linear program. Additionally, we derive a lower bound on the delivery memory trade-off with uncoded placement that accounts for the heterogeneity in cache sizes. We explicitly characterize this trade-off for the case of three end-users, as well as an arbitrary number of end-users when the total memory size at the users is small, and when it is large. Next, we consider a system where the server is connected to the users via rate limited links of different capacities and the server assigns the users' cache sizes subject to a total cache budget. We characterize the optimal cache sizes that minimize the delivery completion time with uncoded placement and linear delivery. In particular, the optimal memory allocation balances between assigning larger cache sizes to users with low capacity links and uniform memory allocation.
△ Less
Submitted 13 March, 2019; v1 submitted 18 October, 2018;
originally announced October 2018.
-
Device-to-Device Secure Coded Caching
Authors:
Ahmed A. Zewail,
Aylin Yener
Abstract:
This paper studies device to device (D2D) coded-caching with information theoretic security guarantees. A broadcast network consisting of a server, which has a library of files, and end users equipped with cache memories, is considered. Information theoretic security guarantees for confidentiality are imposed upon the files. The server populates the end user caches, after which D2D communications…
▽ More
This paper studies device to device (D2D) coded-caching with information theoretic security guarantees. A broadcast network consisting of a server, which has a library of files, and end users equipped with cache memories, is considered. Information theoretic security guarantees for confidentiality are imposed upon the files. The server populates the end user caches, after which D2D communications enable the delivery of the requested files. Accordingly, we require that a user must not have access to files it did not request, i.e., secure caching. First, a centralized coded caching scheme is provided by jointly optimizing the cache placement and delivery policies. Next, a decentralized coded caching scheme is developed that does not require the knowledge of the number of active users during the caching phase. Both schemes utilize non-perfect secret sharing and one-time pad keying, to guarantee secure caching. Furthermore, the proposed schemes provide secure delivery as a side benefit, i.e., any external entity which overhears the transmitted signals during the delivery phase cannot obtain any information about the database files. The proposed schemes provide the achievable upper bound on the minimum delivery sum rate. Lower bounds on the required transmission sum rate are also derived using cut-set arguments indicating the multiplicative gap between the lower and upper bounds. Numerical results indicate that the gap vanishes with increasing memory size. Overall, the work demonstrates the effectiveness of D2D communications in cache-aided systems even when confidentiality constraints are imposed at the participating nodes and against external eavesdroppers.
△ Less
Submitted 2 May, 2019; v1 submitted 18 September, 2018;
originally announced September 2018.
-
On Coded Caching with Heterogeneous Distortion Requirements
Authors:
Abdelrahman M. Ibrahim,
Ahmed A. Zewail,
Aylin Yener
Abstract:
This paper considers heterogeneous coded caching where the users have unequal distortion requirements. The server is connected to the users via an error-free multicast link and designs the users' cache sizes subject to a total memory budget. In particular, in the placement phase, the server jointly designs the users' cache sizes and the cache contents. To serve the users' requests, in the delivery…
▽ More
This paper considers heterogeneous coded caching where the users have unequal distortion requirements. The server is connected to the users via an error-free multicast link and designs the users' cache sizes subject to a total memory budget. In particular, in the placement phase, the server jointly designs the users' cache sizes and the cache contents. To serve the users' requests, in the delivery phase, the server transmits signals that satisfy the users' distortion requirements. An optimization problem with the objective of minimizing the worst-case delivery load subject to the total cache memory budget and users' distortion requirements is formulated. The optimal solution for uncoded placement and linear delivery is characterized explicitly and is shown to exhibit a threshold policy with respect to the total cache memory budget. As a byproduct of the study, a caching scheme for systems with fixed cache sizes that outperforms the state-of-art is presented.
△ Less
Submitted 27 February, 2018;
originally announced February 2018.
-
Combination Networks with or without Secrecy Constraints: The Impact of Caching Relays
Authors:
Ahmed A. Zewail,
Aylin Yener
Abstract:
This paper considers a two-hop network architecture known as a combination network, where a layer of relay nodes connects a server to a set of end users. In particular, a new model is investigated where the intermediate relays employ caches in addition to the end users. First, a new centralized coded caching scheme is developed that utilizes maximum distance separable (MDS) coding, jointly optimiz…
▽ More
This paper considers a two-hop network architecture known as a combination network, where a layer of relay nodes connects a server to a set of end users. In particular, a new model is investigated where the intermediate relays employ caches in addition to the end users. First, a new centralized coded caching scheme is developed that utilizes maximum distance separable (MDS) coding, jointly optimizes cache placement and delivery phase, and enables decomposing the combination network into a set virtual multicast sub-networks. It is shown that if the sum of the memory of an end user and its connected relay nodes is sufficient to store the database, then the server can disengage in the delivery phase and all the end users' requests can be satisfied by the caches in the network. Lower bounds on the normalized delivery load using genie-aided cut-set arguments are presented along with second hop optimality. Next recognizing the information security concerns of coded caching, this new model is studied under three different secrecy settings: 1) secure delivery where we require an external entity must not gain any information about the database files by observing the transmitted signals over the network links, 2) secure caching, where we impose the constraint that end users must not be able to obtain any information about files that they did not request, and 3) both secure delivery and secure caching, simultaneously. We demonstrate how network topology affects the system performance under these secrecy requirements. Finally, we provide numerical results demonstrating the system performance in each of the settings considered.
△ Less
Submitted 13 December, 2017;
originally announced December 2017.
-
Using Network Coding to Achieve the Capacity of Deterministic Relay Networks with Relay Messages
Authors:
Ahmed A. Zewail,
Yahya Mohasseb,
Mohammed Nafie,
Hesham El Gamal
Abstract:
In this paper, we derive the capacity of the deterministic relay networks with relay messages. We consider a network which consists of five nodes, four of which can only communicate via the fifth one. However, the fifth node is not merely a relay as it may exchange private messages with the other network nodes. First, we develop an upper bound on the capacity region based on the notion of a single…
▽ More
In this paper, we derive the capacity of the deterministic relay networks with relay messages. We consider a network which consists of five nodes, four of which can only communicate via the fifth one. However, the fifth node is not merely a relay as it may exchange private messages with the other network nodes. First, we develop an upper bound on the capacity region based on the notion of a single sided genie. In the course of the achievability proof, we also derive the deterministic capacity of a 4-user relay network (without private messages at the relay). The capacity achieving schemes use a combination of two network coding techniques: the Simple Ordering Scheme (SOS) and Detour Schemes (DS). In the SOS, we order the transmitted bits at each user such that the bi-directional messages will be received at the same channel level at the relay, while the basic idea behind the DS is that some parts of the message follow an indirect path to their respective destinations. This paper, therefore, serves to show that user cooperation and network coding can enhance throughput, even when the users are not directly connected to each other.
△ Less
Submitted 26 April, 2014; v1 submitted 18 April, 2014;
originally announced April 2014.
-
The Deterministic Capacity of Relay Networks with Relay Private Messages
Authors:
Ahmed A. Zewail,
Y. Mohasseb,
M. Nafie,
H. EL Gamal
Abstract:
We study the capacity region of a deterministic 4-node network, where 3 nodes can only communicate via the fourth one. However, the fourth node is not merely a relay since it can exchange private messages with all other nodes. This situation resembles the case where a base station relays messages between users and delivers messages between the backbone system and the users. We assume an asymmetric…
▽ More
We study the capacity region of a deterministic 4-node network, where 3 nodes can only communicate via the fourth one. However, the fourth node is not merely a relay since it can exchange private messages with all other nodes. This situation resembles the case where a base station relays messages between users and delivers messages between the backbone system and the users. We assume an asymmetric scenario where the channel between any two nodes is not reciprocal. First, an upper bound on the capacity region is obtained based on the notion of single sided genie. Subsequently, we construct an achievable scheme that achieves this upper bound using a superposition of broadcasting node 4 messages and an achievable "detour" scheme for a reduced 3-user relay network.
△ Less
Submitted 8 August, 2013;
originally announced August 2013.
-
Achievable Degrees of Freedom Region of the MIMO Relay Networks using the Detour Schemes
Authors:
Ahmed A. Zewail,
M. Nafie,
Y. Mohasseb,
H. EL Gamal
Abstract:
In this paper, we study the degrees of freedom (DoF) of the MIMO relay networks. We start with a general Y channel, where each user has $M_i$ antennas and aims to exchange messages with the other two users via a relay equipped with $N$ antennas. Then, we extend our work to a general 4-user MIMO relay network. Unlike most previous work which focused on the total DoF of the network, our aim here is…
▽ More
In this paper, we study the degrees of freedom (DoF) of the MIMO relay networks. We start with a general Y channel, where each user has $M_i$ antennas and aims to exchange messages with the other two users via a relay equipped with $N$ antennas. Then, we extend our work to a general 4-user MIMO relay network. Unlike most previous work which focused on the total DoF of the network, our aim here is to characterize the achievable DoF region as well. We develop an outer bound on the DoF region based on the notion of one sided genie. Then, we define a new achievable region using the Signal Space Alignment (SSA) and the Detour Schemes. Our achievable scheme achieves the upper bound for certain conditions relating $M_i$'s and $N$.
△ Less
Submitted 28 February, 2014; v1 submitted 2 July, 2013;
originally announced July 2013.
-
The Deterministic Multicast Capacity of 4-Node Relay Networks
Authors:
Ahmed A. Zewail,
M. Nafie,
Y. Mohasseb,
H. El Gamal
Abstract:
In this paper, we completely characterize the deterministic capacity region of a four-node relay network with no direct links between the nodes, where each node communicates with the three other nodes via a relay. Towards this end, we develop an upper bound on the deterministic capacity region, based on the notion of a one-sided genie. To establish achievability, we use the detour schemes that ach…
▽ More
In this paper, we completely characterize the deterministic capacity region of a four-node relay network with no direct links between the nodes, where each node communicates with the three other nodes via a relay. Towards this end, we develop an upper bound on the deterministic capacity region, based on the notion of a one-sided genie. To establish achievability, we use the detour schemes that achieve the upper bound by routing specific bits via indirect paths instead of sending them directly.
△ Less
Submitted 16 April, 2013;
originally announced April 2013.
