Coded Computing: A Joint Communication and Computing Design Framework for Tactical Wireless Edge Computing
Abstract
The use of compute-intensive machine learning and big-data analytics for processing a collection of raw data streams from distributed sources is becoming increasingly more prevalent on tactical battlefields in oder to enhance awareness of the environment and act on it. As such, the edge computing architecture has been proposed to better satisfy the service requirements of these emerging applications. Unlike the Cloud computing that stores and processes end-users~ data in remote and centralized datacenters, edge computing brings the provision of services closer to the end-users by pooling the available resources at the edge of the network (e.g., smartphones, tablets,smart cars, base stations and routers). As a result, the main driving vision for edge computing is to leverage the significant amount of dispersed computing resources at the edge of the network to provide much more user-aware, resource-efficient, scalable and low-latency services for IoT.Unlike traditional cloud computing environments, however, tactical battlefield networks arecharacterized by significant challenges with respect to scalability due to (1) severe constraints on bandwidth and other resources including storage and energy; and (2) high dynamics due to mobility and disruptions due to jamming and battle conditions. We propose to develop an innovative framework for large-scale tactical edge computing, named coded computing, which brings new concepts from Shannon~s information theory and coding to overcome these two scalability issues. The proposed Coded Computing architecture essentiallyallows a joint design of communication and computation in edge computing, which is critical for a scalable solution. In particular, coded computing enables to reduce the bandwidth utilization of nodes, at the expense of incurring some additional redundant computation, which is well suited to embedded battlefield devices where the energy of communication is substantially higher thancomputational energy. Furthermore, the proposed coded computing architecture combines coding theory with distributed computing to inject computation redundancy in an unorthodox coded form (as opposed to the state-of-the-art replication approaches) to provide robustness to network dynamics and failures. The key idea here is to encode the data in a way that after computationswe obtain coded redundancy for recovering from delayed, failing, erroneous, or adversarially affected computations. This approach will provide the critical resiliency, security, and privacy that is needed for the tactical edge. Building on our strong preliminary results, in this project we develop a sfoundation for thedesign of coded computing in order to tackle the key bottlenecks that arise in tactical wireless edge computing, namely resource (in particular bandwidth) constraints; high dynamics due to mobility and disruptions in service; and security and privacy of computing in the presence of adversarial and colluding nodes in the network. Furthermore, we also demonstrate the practical capabilities of the developed frameworks via targeted experiments over cloud networks. The project consists of two research Thrusts and 8 carefully designed tasks to accomplish these goals.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 23, 2019
- Source ID
- N000141912372
Entities
People
- Salman A. Avestimehr
Organizations
- Office of Naval Research
- United States Navy
- University of Southern California