Work Package2

WP2: Green Disaster Information Delivery and Rescue Management

Objectives and Expected Results

A main goal of WP2 is to design an architecture for information delivery in disaster scenarios. When a large-scale disaster occurs, providing connectivity between victims in disaster areas and people who are in other areas is a key requirement. Since cellular networks are expected to provide primarily connectivity at the disaster time, WP2 focuses on the two essential problems to make connectivity in cellular networks difficult to achieve: limited capacities of batteries of cellular base stations and fragmented networks due to failures and blackouts.

The concrete objectives of WP2 are the following (as listed in the DoW):

  • Extend the architecture designed in WP1 to support both non-disaster and disaster in a seamless manner and support co-existence with IP, interaction with low-level energy metrics and recipient hierarchies for group communication.
  • Provide support for large-scale energy-efficient disaster information delivery for fragmented/disrupted mobile networks, including routing and cache management algorithm for highly fragmented networks.
  • Develop access control and information management schemes to support message authenticity, terminal authentication/authorization, and privacy-related support, especially in fragmented networks.

The expected results are the following (as listed in the DoW):

  • Framework and analysis of the architecture for information delivery in disaster scenarios.

WP2 success criteria are:

  • The architecture successfully satisfies the societal requirements defined in Task 1.1 on the energy efficiency and survivability of the information delivery infrastructure in disaster.
  • At Least 40% Reduction of Power Consumption of GreenICN for Disasters.

In the following, the work that has been done in each respective task towards the aforementioned objectives is described, as well as key achievements that have been accomplished.

Task 2.1: Framework

Summary of Y1

The following work was done and the following results/achievements were obtained:

  • Several detailed use cases as well as requirements from the perspective of WP2 (i.e. disaster scenarios ) have been developed and were contributed to WP1.
  • Concrete architectural requirements were specified and principle approaches to fulfil these requirements were investigated. These results were also contributed to WP1 for the overall design of the GreenICN architecture.
  • A detailed reference scenario for disaster use cases was developed.
  • A first approach for message aggregation (to achieve energy savings) was proposed and studied.
  • An analysis framework for wireless protocol performances was developed, which enables to evaluate various solutions in fragmented networks.

 

Summary of Y2

We created two network scenarios to clarify the WP2 success criteria. 95% connectivity in fragmented networks applied to physical damaged area and 40% power reduction applied to power loss area. We studied two technologies, collaborative communications and cell zooming, for reduction of power consumption for disasters.

In addition, we lead all the tasks 2.1, 2.2 and 2.3 especially in terms of energy reduction.  We developed an energy consumption model for a LTE-based BS. The model is designed to analytically model how frequency resources are used for various radio propagation qualities. It will be used to calculate energy consumed for use scenario in all tasks and how collaborative communication (which was called message aggregation in Year2) reduces energy consumed by a BS is evaluated in Year 2.

Further, we designed a protocol which seamlessly integrates a Publish/Subscribe mechanism, i.e., COPSS, and collaborative communication in order to deliver crucial information to multiple recipients in shelters of a black out area in an energy efficient manner. This protocol is designed so as to interwork with a LTE-based cellular network. How energy is reduced in ideal cases wherein refugees do not move is analytically calculated assuming a shelter wherein many users reside in a small area. Further, to enable efficient group communication mechanisms for efficient rescue operations, we presented the design of policy/role-based forwarding scheme that can prioritize ICN interests depending on the role and priority level.

Finally, we refined a use scenario at black area in terms of how refugees are distributed in such an area. Under such a scenario, how switched-off BSs should be scheduled to maximize the total size of areas covered by active BSs is analyzed by formulating cell zooming as an integer linear programming problem. Then, to solve the problem in feasible time, we design a greedy algorithm which chooses the most efficient cell configuration.

Summary of Y3

The paper “The Benefit of Information Centric Networking for Enabling Communications in Disaster Scenarios” (which has been accepted and been presented at the IEEE Globecom 2015 Workshop on Information Centric Networking Solutions for Real World Applications (ICNSRA)) was completed and published. This work can be regarded as a somewhat position paper from the project on why and how ICN technologies can be useful to tackle the typical communication problems in disaster scenarios. It provides an overview on the challenges, outlines why ICN generally is useful, and gives concrete solution examples that have been developed in GreenICN WP2. Further, the assumptions on energy have been clarified and updated. Also, a paper on how collaborative communication reduces consumed frequency resources has been published (LANMAN 2015), and the proposed greedy algorithm for cell zooming has been revised. A main contribution about the collaborative communication is that frequency resource reductions are validated by experimentally measuring how commercial cellular devices use frequency resources depending on their radio signal qualities. The experiments show that the collaborative communication has potential to reduce energy consumption of BSs (Base Stations) about 30 %.

Task 2.2: Support Routing and caching in fragmented networks

Summary of Y1

The following work was done and the following results/achievements were obtained:

  • Mechanisms to deliver local disaster management information to mobile devices in fragmented networks without the help of any central mobility management node have been designed. Routing and mobility support based on mules has been studied, which are equipped with storage space and move around the disaster-stricken area gathering information to be disseminated. A safety confirmation delivery scheme for supporting mobility and routing based on the COPSS architecture has been proposed. the effectiveness of the scheme in terms of message reachability was analysed, and it was found under what conditions 95% message reachability in fragmented networks can be achieved.
  • A naming scheme to distinguish and differentiate between high-priority and low-priority messages and deliver high-priority messages first has been designed and evaluated. Prioritisation is not used only to support faster delivery of high-importance messages, but also to avoid messages being starved because of nodes running out of battery. The design is called “ame-ased replication” (NREP) and as the name suggests, it makes decisions on whether to replicate messages or not based on the information exposed in the content name. NREP targets post-disaster infrastructureless environments, where first responder teams need to inform the general public of the situation, as well as the management associated with it.

Summary of Y2

In Year 2, we have added a new routing table in the CCN architecture in order to be able to maintain and resolve information when the network gets fragmented. Although originally we were planning to design a naming scheme for that purpose, our efforts led to the modification of the routing machine itself. Our previous work on “Name-based Replication” has dealt to an extent with names for under-delivered information.

We further investigated the performance of mule-based communication under network fragmentation and given and ICN environment. In doing this, we have experimented with several different mobility models, both fixed and random. We have also designed an algorithm to estimate the popularity of interests in a fragmented network and aggregate requests according to that estimation/indication. This mechanism is expected to improve performance considerably.

Routing and relaying was also considered under the newly proposed DID framework, which is also taking into account priorities for certain data items. This group of studies completes the tasks promised during the previous year, with the exception of further studies of NREP and the related tradeoffs mentioned in D2.2.1.

Summary of Y3

A main contribution has been the further revised and improved “Information Resilience” scheme already proposed in Year 2 and presented at IFIP Networking 2015.  Based on the so-called SIT table, we exploit off-path caches, as well as user caches to retrieve content, when the path to the main source has failed (e.g., after network fragmentation). A significantly improved performance evaluation of this scheme with new caching strategies and an extended set of evaluation scenarios has been conducted.

Task 2.3 Access Control and Management in fragmented networks

Summary of Y1

The following work was done and the following results/achievements were obtained:

  • Access control to specific information items was studied. Data-Centric Security techniques were investigated, namely “Ciphertext-Policy Attribute Based Encryption” (CP-ABE), where security and privacy rely on information contained in the message itself. This ability is particularly important in the fragmented network scenarios tackled by GreenICN. Also, a study was performed to evaluate different naming and digital signature schemes, in terms of their verification time and overhead. Further, procedures to verify content validity in network caches were studied.
  • Access control regarding the control of devices / terminals was studied. In this field, decentralized technologies for a seamless authentication and authorization before and after a disaster were investigated. The exploitation of Hierarchical Identity Based Encryption (IBE) and Identity Based Signatures (IBS) in ICN was studied by building on the work of 3GPP and Wi-Fi Alliance on new standards to establish a Point-to-Point link between adjacent nodes, able to operate even when the network infrastructure is down. Further, a login procedure for Facebook based on Hierarchical IBS was evaluated, by showing that the user experience is not affected by using Hierarchical IBS/IBE instead of current Internet services.
  • A naming scheme for authentication of messages in decentralised disaster scenarios was proposed. The approach uses a Web-of-Trust in conjunction with self-certifying names for the fragmented (mobile) networks scenario, where connectivity to centralised entities and authentication servers is not available.
  • The energy efficiency of the proposed IBS/IBE terminal/network authentication procedure was compared with that of SE-EPS-AKA (Authentication and Key Agreement), the advanced version of the current 3GPP AKA.

 

Summary of Y2

We investigated security mechanisms depending on network status, meaning solutions that adapt procedures and performances to the status of the network; for instance, after a disaster some pre-defined encrypted information could be released; the challenge is in designing distributed solutions that do not necessarily depend on interactions with central servers/authorities that may be unavailable/unreachable.

Further, we extended IBS to be used in a moderator-controlled information sharing service such as NetNews. We proposed Identity-Based Aggregate Signatures (IBAS) as yet another signature scheme for ICN (specifically for NDN), which reduces the signature overhead.

Furthermore, our approach for decentralized authentication has been further developed, generalized, and thoroughly evaluated: a) Analytical results have been obtained, b) we developed a methodology for artificially generating Web-of-Trust graphs, c) The basic scheme has been generalized, applied to more use cases, and been thoroughly evaluated by means of a prototype implementation, and d) we evaluated the scalability and performance our scheme on a prototype implementation on actual PGP WoTs as well as on large-scale artificially generated WoT graphs.

Finally, a comparison of energy efficiency of the proposed vs. existing authentication procedures was further investigated and the difficulties have been much better put into focus.

Summary of Y3

The techniques to access more “dynamic” resources (e.g. a web-page, a database, etc.) provided by emergency teams to support operations in disaster scenario have been designed. The access to such kind of resources cannot be easily protected by using the previous mentioned “un-managed” access-control techniques and calls for a “managed” access-control. In particular, we examined solutions based on anonymous credentials that enable to implement a CP-ABE like attribute-based access control and results to be very suitable for accessing dynamic resources. Further, regarding Access control of terminals and services, we configured MIS equipped with IBAS using several PC’s as IBAS nodes and monitored the performance. IBAS was shown to reduce the overhead of signature for safety confirmation about 45% to 60% compared to RSA-2048 and has almost the same throughput as that of RSA-2048. Finally, regarding energy, the actual energy consumption of IBAS is calculated based on a simple experiment. The results are affirmative to IBAS. In-network caches reduces 30% of the energy consumption of IBAS to implement SNS in total.