In recent years, the advent of agile RF transceivers, software defined radios, and large Field-Programmable Gate Array (FPGA) devices has both decreased the timescale for development and deployment of communications systems and increased the rate of emergence of threats resulting in the need to rapidly adapt. To shorten the timescales of inserting new technology into tactical communications systems, the Common-modem Hardware Integrated Library (CHIL) was developed. CHIL is a library of configurable digital signal processing blocks grouped together in a framework that enables waveform developers to dynamically “string” processing blocks together to form new waveforms. While CHIL has the potential to offer significant benefits to waveform development, it suffers from several limitations including: 1) tight coupling of the framework with library algorithms, 2) slow operation and large FPGA utilization, and 3) lack of documentation. In this paper, we present enhancements to CHIL to address some of these limitations and demonstrate increased reusability, reduction in FPGA utilization, and increase in operating speed.

Recent popularity of agile radio frequency transceivers, software defined radios, and large Field-Programmable Gate Array (FPGA) devices presents both opportunities and challenges in the tactical communications domain. Technology advances enable more complex communications systems, but also allow increasingly sophisticated threats to existing systems at faster rates. Traditional, stove-piped, special-purpose tactical waveform development techniques are not suited for this new environment, creating a clear need for rapid waveform design and deployment practices. The maturation of modular, evolvable design frameworks offers the flexibility to refresh system capabilities without requiring entire system overhauls. GNU Radio and Common-modern Hardware Integrated Library (CHlL) are two such frameworks that are intended for General Purpose Processor (GPP) and FPGA/ASIC deployment, respectively. In this paper, we outline an integrated approach for prototyping a surrogate tactical waveform using both the CHlL and GNU Radio frameworks, provide benchmarking results to highlight the implementation's configurability, and comment on lessons learned while interfacing both frameworks together.

In recent years, there has been a large push in the U.S. Department of Defense (DoD) to more rapidly respond and adapt to changing technology advancements and emerging communications systems threats. One issue has been that traditional DoD waveform development has been stove-piped in that processing blocks are implemented to serve a single function instead of made generic and configurable to support multiple waveforms. Developing new waveforms typically requires starting from scratch almost everytime. In this paper, we present Common Hardware-modem Integrated Library (CHIL). CHIL is both a library of configurable processing blocks as well as a framework that ties these blocks together to rapidly instantiate new DoD and non-DoD waveforms. We present an overview of the CHIL framework and various components as well as provide a test-case where CHIL was leveraged to instantiate a DoD waveform.

Wireless network protocol research typically requires evaluating performance over a set of controlled wireless link conditions. Although ns-3 provides wireless models like WiFi and WiMax, they have dozens of parameters that affect performance and it is difficult to control link rate, latency, error, and other attributes. To mitigate this issue, we extended a basic, range-based ns-3 model called SimpleWireless. Features that were added to this model include transmission delay, configurable queues that enforce a data rate, support for differentiation of control and data traffic, support for several configurable error models, support for directional networking, support for fixed contention and finally, support for PCAP packet capture. The goal of these additional features is to provide network protocol researchers with a basic yet feature rich wireless model that enables evaluating their protocol in a controlled wireless environment. In this paper, we describe the base SimpleWireless model and each feature that has been developed to enhance that model and create the so called "LL SimpleWireless" model. Additionally, we provide information on performance evaluation of the LL SimpleWireless model to verify functionality of the added features.

Airborne tactical networks (ATNs) have provided protected air-to-air communications for military aircraft for several decades. To support emerging and future warfighter needs, the next generation of systems will require significant improvements to provide higher capacity, longer range, greater flexibility, and increased interoperability. Governed by domain characteristics such as long transmission ranges, low-to-medium data rates, latency constraints, and link protection needs, the air tactical domain poses several unique requirements on link and network design. Developing next-generation ATNs requires an understanding of the airborne tactical domain, including the design constraints and challenges at various layers of the network stack. In this article, we provide an overview of the unique domain characteristics of ATNs and highlight the key design challenges and research areas associated with the physical, link, and network layers.

In recent years, there's been a large push in the U.S. Department of Defense to provide greater bandwidth efficiency amidst congested spectrum. This is especially true in airborne tactical networks where transmit rates are limited and optimized for interference mitigation. Airborne tactical networks (ATNs) differ from many ground networks in that most of the traffic is broadcast and multicast. One method to more efficiently utilize the medium is to build a connected dominating set (CDS) backbone to reduce relay redundancy with overlapping nodes. While building a CDS has been shown to reduce retransmissions, thus saving bandwidth, backbone nodes can easily become congested. In this paper, we apply network coding to backbone CDS in airborne tactical networks to potentially mitigate the issues of congestion. Specifically, we implement network coded CDS (NCDS) in the Linux kernel and evaluate its performance compared to no coding under various relevant topologies. The results show that although NCDS can provide gains in tightly controlled topologies, the gains are severely limited in random and relevant ATN topologies.

In recent years, there has been increasing interest in the US Department of Defense to build an on-demand airborne network for communications relay utilizing high-capacity, long-range military radio systems. While these systems operate well in a network of homogeneous systems, platforms generally employ multiple heterogeneous radio systems making internetworking difficult due to varying radio characteristics and lack of interoperability. Although simulations and emulation tests can provide a baseline for how systems will perform in a controlled environment, field tests are crucial to demonstrate capabilities in real-world operating environments. In this paper, we present measurement results from a field test involving two airborne platforms forming a dynamically routed aerial IP backbone over 200 nautical miles with various radio systems as part of the C4ISR on-the-move 2010 exercise. We present measurement results on per link performance, radio-to-router interface performance, and multihop network performance results with prototype software on open source platforms. Additionally, key lessons learned and recommendations are given.

The explosion of net-centric applications in the DoD tactical edge has increased pressure on DoD communications systems to maximize bandwidth efficiency amidst a shrinking allocation of RF spectrum. Coupled with the need for reliable transmissions at the tactical edge, non-standard techniques are increasingly being assessed for their suitability in military networks. One such method is applying Network Coding to provide bandwidth efficiency and increased reliability by sending linear combinations of packet data across multiple paths. While there have been many userspace network coding implementations available in the open source community and in academia, there are no known Linux kernel broadcast network coding implementations. Because the majority of network functionality is in the kernel, a network coding kernel implementation would greatly speed up packet processing, reduce needless userspace copying, and allow researchers to study its effects in real-world environments. In this paper, we present a Linux kernel implementation of inter-flow network coding using XOR for broadcast traffic over broadcast links. In addition to the design and layout, we provide a basic functional evaluation to demonstrate its effectiveness in reducing bandwidth requirements.

In highly dynamic wireless environments, link metrics such as link quality, availability, and others have become increasingly important to enable smart multi-hop routing decisions. In recent years, a number of radio-to-router interface (R2RI) protocols such as Point-to-Point over Ethernet RFC5578, Dynamic Link Exchange Protocol (DLEP), and Radio-Router Control Protocol (R2CP) have emerged to address the need to have a common set of link metrics exposed from the radio to the router to enhance multi-hop routing decisions. In the past, testing R2RI required separate processes and proxies to convert layer 2 link metrics into R2RI standard formats. Although useful for allowing legacy radio systems to comply with R2RI formats, the proxy technique is difficult for building large-scale (100+ node) test environments with emulated radios and routers. In this paper, we describe modifications made to a popular open source link emulator, EMANE, to support emerging radio-to-router interface (R2RI) standards, DLEP and R2CP, on several EMANE MAC models including 802.11 and CommEffect. Additionally, a functional evaluation of the system is performed1.

Military radio systems are often constrained systems that require per link information to effectively institute QoS policies, route packets, manage topology, and tunnel application data. One of the first attempts at providing standard radio-to-router information through a common interface was point-to-point over Ethernet (PPPoE) RFC5578. Although PPPoE RFC5578 works effectively for directional radios that form point-to-point links, there are a number of limitations including added overhead, requirement of Ethernet as the medium between radio and router, and inefficient medium reuse in broadcast environments. As a result, many new protocols such as Dynamic Link Exchange Protocol (DLEP) and Radio-to-Router Control Protocol (R2CP) have emerged to address the limitations of RFC5578 and provide a standard method to share per link information with the network layer. In previous work, an open source routing solution using a modified Quagga router to support dynamic link metrics were developed to support an open source implementation of radio-aware routing with RFC5578. In this paper, we present significant changes to this system to support both DLEP and R2CP on the open source router (OSR). These changes facilitate open source support of all 3 radio-to-router protocols and enable performance comparisons.

In highly dynamic wireless environments, link metrics such as link quality, availability, and others have become increasingly important to enable smart multi-hop routing decisions. In recent years, a number of radio-to-router interface (R2RI) protocols such as Point-to-Point over Ethernet RFC5578, Dynamic Link Exchange Protocol (DLEP), and Radio-Router Control Protocol (R2CP) have emerged to address the need to have a common set of link metrics exposed from the radio to the router to enhance multi-hop routing decisions. To fully evaluate R2RI functionality and specifications, differing implementations of both radio/client and router/server-side R2RI protocols must be prototyped and tested. In this paper, we present comparison tests of each of the three (3) radio-to-router interfaces with two (2) router/server-side R2RI experimental/beta R2RI implementations: one on a commercial Cisco router, and one on an open source Quagga platform. The goal of the comparison is not necessarily to provide a holistic performance comparison (as much of the code is experimental), but to highlight implementation differences and potential issues. In many cases, issues are already resolved in future releases.1

Tactical wireless and mobile networks are the primary networking infrastructure in the Global Information Grid (GIG) to provide end-to-end connectivity to the warfighters at the tactical edge. The highly dynamic nature of tactical edge networks raise a number of challenging issues related to data transport and service delivery in the tactical environment. To address some of these issues, DoD waveforms have increasingly leveraged layer 2 link information to make smart cross-layer multihop routing decisions. Although there has been some measure of success in providing higher end-to-end data delivery, the lack of standard interfaces between the radio and router have led to interoperability issues in environments with a heterogeneous mix of radio systems. As a result, there has been increased desire to standardize radio-to-router interfaces (R2RIs) as a means to separate radio and router functionality and to allow greater interoperability between systems. In this article, we examine three R2RI protocols currently being vetted through the Internet Engineering Task Force and currently integrated or under consideration in DoD radio systems (RFC 5578, R2CP, and DLEP), and identify their current use and applicability in the tactical edge. Furthermore, we identify some challenges in implementing any R2RI scheme into emerging systems.

The current generation of long-range, high capacity, military radios are stove-piped systems that work well in a homogeneous environment, but require significant setup and configuration to interoperate with other radio systems. Each radio provides a subset of disparate link information in nonstandard interfaces and has built-in home-grown or industry-based routers running potentially different routing protocols. In a heterogeneous radio system airborne environment, wireless link characteristics change rapidly, often requiring direct link feedback from the radio to make routing decisions. In recent years, there has been a number of work in developing a common radio-to-router interface that standardizes a subset of per-link information to pass to the network layer for use in dynamic MANET routing. While simulations and emulation tests can provide a baseline for how systems will perform, field-tests are crucial to demonstrate capabilities in real-world operating environments. In this paper, we present measurement results from a field test involving three airborne and two ground assets with various radio systems that test an implementation of RFC4938, a radio-to-router interface protocol, and its interaction with a modified OSPFv3 routing protocol to support dynamic link metrics and OSPF cost generation. The assets participated in the exercise formed a high capacity, dynamically routed aerial IP backbone made of heterogeneous radio technologies over 250 nautical miles (Nm), allowing the passing of military operational traffic.1

In highly dynamic airborne environments, per link information becomes crucial in effectively routing packets throughout the network. Point-to-point Protocol over Ethernet (PPPoE) RFC4938/55781 describes an elegant way to standardize an interface to transmit per link information to the router to make educated routing decisions. These dynamically changing point-to-point links to the router pose an interesting challenge in defining interfaces in the router configuration and providing multicast/broadcast emulation. In this paper, we present an open-source, Linux implementation of a Common Virtual Multipoint Interface (CVMI). The CVMI aggregates multiple point-to-point interfaces into a virtual interface and provides multicast/broadcast emulation on these dynamically changing interfaces. We describe our implementation in detail and show how each link can be grouped to perform differently despite the homogeneous nature of PPP links and provide some basic performance evaluations to show functionality.

Airborne network (AN) backbones hold the promise of providing the persistent high-bandwidth line-of-sight (LOS) conduit that interconnects the various mission elements in a battlespace. When properly designed the control of such a backbone should make the collection of links that form the backbone appear, as much as possible, as a seamless high-bandwidth connection between mission elements communicating through it. This can be accomplished by dynamically adapting the topology of the backbone network in a manner that takes into account flow demands and communications link capacities across the entire network, and by applying quality-of-service (QoS) techniques, to attain the desired performance. This goal must be satisfied despite the fact that connectivity is intermittent, nodes leave and join the network frequently, and a diverse set of communications technologies is used on the nodes that form the backbone. Accomplishing this task requires the use of control information distributed across various network layers. We use a Network Control Plane (NCP) for this purpose. A communications architecture designed specifically for the Airborne Network problem has been prototyped and characterized using simulation and emulation. During flight tests conducted in August of 2006, many of the prototyped concepts were verified in a 3-node airborne network backbone based on directional RF links. In this paper we describe the infrastructure evaluation portion of the flight tests and implementation of the prototyped AN architecture in the 3-node backbone, with emphasis on the Network Control Plane. Measured end-to-end performance obtained from data gathered during the experiment is presented.

The Airborne Network Definition (AND) project had the goal of creating and testing a robust, efficient network architecture networking all elements of the battlefield. This effort has since been generalized to fit the goals of the Mobile Edge Network System Architecture (MENSA) effort. The network is designed to be self-contained, attaching to fixed backbone infrastructure whenever possible. The fundamental building block of the architecture is the Small Combat Network (SCN), which integrates heterogeneous ground and air platforms, facilitating collaborative applications such as Blue Force Tracking, Cooperative Sensing and Targeting. The architecture uses the concept of an IP core to network unlike domains (radio types) and to connect the SCN to the backbone and the Global Information Grid (GIG). This paper describes the requirements of the network and outlines the technical design of the SCN architecture. We present step-by-step descriptions of a communication on the SCN which highlights some of the key features of the architecture. We present results of a simulation that applies our proposed architecture to realistic warfighting scenarios. Results show that the architecture enables cooperative applications and point to future work that will design and evaluate a deployable AN network architecture.

In this paper we consider the problem of autoconfiguring nodes in an airborne wireless network. The airborne platforms consist of routers as well as one or more hosts. External links from the airborne platform employ directional transmit beams. We expect the nodes to join or leave the network without pre-planning or manual configuration. Additionally we expect the users and applications of this network to discover available services in an automated manner. In this paper we present protocols that allow autoconfiguration of IPv6 addresses and names, as well as the automated discovery of services within the airborne wireless network.

Future military networks will include IP connectivity among mobile airborne platforms. To understand the complexities of such networks, MIT Lincoln Laboratory (MITLL) through a program known as Paul Revere (PR), has implemented a multi-link, IP-networking system between a Boeing 707 and 45 foot tractor trailer. As an experimental platform, the system also supported operator and application requirements during the Joint Expeditionary Force Experiments (JEFX) 2004 executed by the United States Air Force. In this document, we motivate and describe the architecture and implementation, present analysis of our findings, and provide general architectural guidelines for the construction of future IP-based systems.