Efficient and Reliable Car-to-Cloud Data Transfer Empowered by BBR-Enabled Network Coding

Random linear network coding (RLNC) has the potential to improve the performance of current and future Internet of Things (IoT) communication systems, but is computationally demanding due to matrix multiplications and inversions. Some single-core RLNC implementations achieve already sufficient coding speeds for contemporary multimedia streaming formats. However, advances in multimedia streaming formats and IoT applications will require the exploitation of heterogeneous multicore architectures, which are becoming common for a wide range of IoT nodes, including smartphones. In this paper, we introduce and evaluate efficient RLNC computing strategies for IoT node architectures, including the emerging heterogeneous big.LITTLE multicore architectures with multiple big (fast) cores and multiple LITTLE (slow) cores. In contrast to existing RLNC implementation strategies, we build on and adapt highly optimized dense matrix operations from the high performance computing field to RLNC on heterogeneous multicore IoT nodes. Our approach includes the optimization of RLNC matrix operations through optimized operations on matrix blocks with single instruction multiple data instructions. We schedule block operations on the heterogeneous cores through a directed acyclic graph that avoids artificial synchronization points while ensuring the data dependencies. We examine priority scheduling according to the number of outgoing dependencies of a task and data locality of cached blocks. Our extensive measurements with several heterogeneous big.LITTLE multicore IoT node and smartphone processor boards demonstrate higher RLNC encoding and decoding throughputs than existing approaches. Moreover, our measurements indicate that the utilization of more cores decreases energy consumption, which is an important goal for IoT nodes.

Random linear network coding with ladder-shaped global coding matrix for robust video transmission

The streaming of real-time audio/video data is very challenging because of the time-varying and unreliable wireless channels, video content characteristics, limited bandwidth, dynamic topology, heterogeneous and distributed environment and high packet loss rate because of wireless interference and channel fading. Because of such open issues present in wireless networks, it is very difficult to satisfy the requirements of audio/video streaming applications such as low delay, low packet loss, jitter control etc. In such scenarios of opportunistic networks where real time data is being streamed from server node to client node, Network Coding and its variants can be used by such nodes to meet different requirements. Network coding changes the role of such multimedia nodes from store and forward to encode data packets. Encoding process includes mathematical operations on data packets. This survey paper has considered number of speculative and practical approaches and cases where network coding or its variant applied either fully or partially on multimedia traffic with the aim to improve performance and to provide protection against packet losses. This review has mainly focused on the opportunity of performance enhancement of MPEG-4 traffic over Wireless Network using Random Linear Network Coding (RLNC) with Multi Generation Mixing (MGM). Using Multi Generation Mixing, packets of greater importance has got more protection, less loss, more reconstruction and recovery of real time data.

Wireless networks typically employ some form of forward error correction (FEC) coding and some automatic repeat request (ARQ) protocol to ensure reliable data transmission over lossy channels. We propose to integrate FEC and ARQ in the context of random linear network coding (RLNC). In particular, we develop Caterpillar RLNC with feedback (CRLNC-FB), an RLNC approach with a finite sliding packet transmission window in conjunction with feedback-based selective repeat ARQ. CRLNC-FB employs a novel RLNC decoding method based on a band-form of Gaussian elimination. In response to lost packets, CRLNC-FB retransmits lost packets in systematic (uncoded) form to aid fast in-order packet delivery at the receiver. Extensive performance evaluations indicate that CRLNC-FB gives higher throughput-delay performance than the preceding RLNC approaches with feedback. In particular, CRLNC-FB with its sliding window achieves lower delays than block-based RLNC. Also, the retransmission of uncoded source packets in CRLNC-FB contributes to a significantly higher throughput-delay performance than loss recovery through coded packets interspersed among future source packets at a prescribed code rate.

The network coding capacity of a single multicast traffic is characterized by the min-cut/max-flow (mcMF) theorem, which can be achieved by random linear network coding (RLNC). Nonetheless, the graph-theoretic characterization for multiple unicast/multicast traffic remains an open problem. This paper proposes and studies a new class of intersession-network-coding schemes: RLNC with selective cancelling (SC), which inherits the complexity advantage of RLNC once the set of selective cancelling edges is decided. A graph-theoretic characterization is provided for the achievable rates of RLNC with SC for the general multiple multicast setting. The findings contain most existing achievability results as special cases, including the mcMF theorem of the single multicast traffic and the existing characterization of pairwise intersession network coding. One prominent feature of the proposed approach is its focus on the achievability analysis with arbitrary network topology, arbitrary inter-session packet-mixing capability, and arbitrary traffic demands, which distinguishes the results from the special case analysis, capacity outer bound constructions, and the pattern-based (butterfly-based) superposition arguments.

Passive network tomography uses end-to-end observations of network communication to characterize the network, for instance to estimate the network topology and to localize random or adversarial glitches. Under the setting of linear network coding this work provides a comprehensive study of passive network tomography in the presence of network (random or adversarial) glitches. To be concrete, this work is developed along two directions: 1. Tomographic upper and lower bounds (i.e., the most adverse conditions in each problem setting under which network tomography is possible, and corresponding schemes (computationally efficient, if possible) that achieve this performance) are presented for random linear network coding (RLNC). We consider RLNC designed with common randomness, i.e., the receiver knows the random code-books all nodes. (To justify this, we show an upper bound for the problem of topology estimation in networks using RLNC without common randomness.) In this setting we present the first set of algorithms that characterize the network topology exactly. Our algorithm for topology estimation with random network errors has time complexity that is polynomial in network parameters. For the problem of network error localization given the topology information, we present the first computationally tractable algorithm to localize random errors, and prove it is computationally intractable to localize adversarial errors. 2. New network coding schemes are designed that improve the tomographic performance of RLNC while maintaining the desirable low-complexity, throughput-optimal, distributed linear network coding properties of RLNC. In particular, we design network codes based on Reed-Solomon codes so that a maximal number of adversarial errors can be localized in a computationally efficient manner even without the information of network topology.

Network Coding (NC) for Wireless Mesh Network (WMN) has received a lot of attention from the research community in recent years due to its ability to provide higher throughput, reliability, and efficiency in the use of the available wireless spectrum. NC has had two main research lines focused on inter-flow and intra-flow coding. However, these two domains have historically been studied separately. This paper proposes a testbed implementation of CORE, the first protocol to bridge both approaches such that the structure of intra-flow coding can be exploited during inter-flow coding to both enhance reliability, common of the former, while maintaining an efficient spectrum usage, typical of the latter. This paper uses the intuition provided in [1] to propose a practical implementation of the protocol leveraging Random Linear Network Coding (RLNC) for intra-flow coding, a credit based packet transmission approach to decide how much and when to send redundancy in the network, and a minimalistic feedback mechanism to guarantee delivery of generations of the different flows. Given the delay constraints of video applications, we proposed a simple yet effective coding mechanism, Block Coding On The Fly (BCFly), that allows a block encoder to be fed on-the-fly, thus reducing the delay to accumulate enough packets that is introduced by typical generation based NC techniques. Our measurements and comparison to forwarding and COPE show that CORE not only outperforms these schemes even for small packet losses, but that the gains can be of five fold and two orders of magnitude, respectively, for higher packet losses.

A typical intelligent factory, a smart home or a smart city consists of several nodes or machines. In a typical Machine to Machine Communication mode (M2M), each machine senses some information (pressure, temperature, presence of person, etc.) and then communicates this information to other machines. This communication could be unicast or multicast depending on the application. Further, the destination node of the M2M communication is many-a-times not in the direct communication range. In the presence of tens and hundreds of machines forming a network, it is essential to have an efficient communication mechanism for multihop multicast transmission between different machines. This paper proposes a network coding (NC) based multihop multicast communication for different topologies - butterfly, flattened butterfly, hypercube and folded-clos model. It is observed that random linear network coding (RLNC) technique outperforms best routing protocol for different topologies, by requiring less time for data transmission, reducing the time required by upto 50%, depending on the model. Further, in order to improve the performance of network coding for M2M in a disconnected network; an integrated network coding and routing approach is proposed and evaluated. It is observed that such an integrated mechanism requires upto 25% less time than pure routing and upto 10% less time than required by RLNC.

Network error localizations uses end-to-end observations to localize random or adversarial glitches. For error localization under random linear network codes (RLNCs) the schemes proposed in previous literatures require the priori knowledge of the network topology. Moreover, adversarial error localization is proved computationally intractable under RLNCs. The goal of current paper is designing new network coding schemes to improve the error localization performance of RLNCs while maintaining the key advantages of RLNCs. To be concrete, we introduce Network Reed-Solomon codes (NRSCs), which have the following features: 1. NRSCs are low-complexity distributed linear network codes. 2. NRSCs achieve the network multicast capacity with high probability. 3. In adversarially faulty networks NRSCs enable the receiver to locate a maximum number of adversarial errors in a computationally efficient manner. Moreover, the error localization schemes under NRSCs do not need the priori knowledge of the network topology and thus are robust against dynamic network updating.

In a network application, channel coding alone is not sufficient to reliably transmit a message of finite length K from a source to one or more destinations as in, e.g., file transfer. To ensure that no data is lost, it must be combined with rateless erasure correcting schemes on a higher layer, such as a time-division multiple access (TDMA) system paired with automatic repeat request (ARQ) or random linear network coding (RLNC). We consider binary channel coding on a binary symmetric channel (BSC) and q-ary RLNC for erasure correction in a star network, where Y sources send messages to each other with the help of a central relay. In this scenario RLNC has been shown to have a throughput advantage over TDMA schemes as K→∞ and q→∞. In this paper we focus on finite block lengths and compare the expected throughputs of RLNC and TDMA. For a total message length of K bits, which can be subdivided into blocks of smaller size prior to channel coding, we obtain the channel code rate and the number of blocks that maximize the expected throughput of both RLNC and TDMA, and we find that TDMA is more throughput-efficient for small message lengths K and small q.

A fundamental understanding of the relationship between delay performance and complexity in network coding is instrumental towards its application in practical systems. The main argument against delay-optimal random linear network coding (RLNC) is its decoding complexity, which is O(n 3 ) for n original packets. Fountain codes, such as LT and Raptor codes, reduce the decoding load on the receiver but at the cost of introducing additional, non-negligible delay. The source of this increased delay is the inherent sparsity of the code, which significantly reduces the impact of a new coded packet, i.e., the probability of a packet to be linearly independent from the knowledge at the receiver, as we approach the end of the transmission (when few degrees of freedom are missing). Thus, the additional overhead is mainly due to the transmission of the last data packets. Our key observation is that switching gears to denser codes as the transmission process progresses could considerably reduce the delay overhead while maintaining the complexity advantages of a sparse code. We propose tunable sparse network coding as a dynamic coding mechanism with a code structure with two regions: a sparse region, with various levels of sparsity, and a dense region, where packets are generated as per RLNC. We characterize the problem in multicast sessions on general networks and illustrate trade-offs in especial cases of interest. We also present a novel mechanism to perform efficient Gaussian elimination for sparse matrices that guarantees linear decoding complexity with high probability.

Network coding is a promising technology which allows coding of multiple packets in routers. Random linear network coding (RLNC), a distributed network coding scheme, insists each node to forward independent random linear combinations of incoming packets through its outgoing links. RLNC appends a header which contains the information about how the linear combinations are taken. We observe that some additional flexibility can be induced in RLNC header with minor modification in decoding. Our modification does not make any increase in computational complexity. We also demonstrate some applications of it in secure network coding and error correction.

In this paper we propose random linear network coding (RLNC) as an efficient encoding scheme for unicast flows in single-hop ad hoc network and multi hop wireless mesh networks (WMN). The traditional approach for encoding unicast flows, known as inter-flow network coding (NC), assumes instant decoding and assumption of reliable feedback channel to minimizes the total number of transmissions. But under the practical consideration of the wireless medium access control (MAC), the gain of inter-flow NC technique suffers from higher latency due to the MAC overhead. We validate the result of our work on TelosB motes testbed. Our testbed implementation shows that the proposed approach of using RLNC reduces the transmission completion time by up to 40% compared to the inter-flow NC scheme for single-hop data dissemination and multi-hop wireless routing problems.

네트워크 코딩(Network Coding)은 통신망의 성능 향상에 도움을 줄 수 있으나 이의 소프트웨어적 구현은 부호화/복호화 단계에서 매우 큰 지연시간을 유발할 수 있어 이를 줄일 수 있는 병렬화된 구현이 필수적이라 할 수 있다. 본 논문에서는 랜덤 리니어 네트워크 코딩(Random Linear Network Coding)과 랜덤 리니어 네트워크 코딩의 단점을 보완하고자 최근 제안된 파이프라인 네트워크 코딩(Pipeline Network Coding)의 병렬처리 성능을 비교한다. 또한, 네트워크 코딩의 CPU에서의 병렬처리 기법과 GPGPU(General Purpose Graphics Processing Units)에서의 병렬처리 기법을 비교하여 네트워크 코딩의 사용 시 그 파라미터에 따라 적절한 병렬처리 기법을 선택할 필요성이 있음을 보여준다. Network coding has been shown to improve various performance metrics in network systems. However, if network coding is implemented as software a huge time delay may be incurred at encoding/decoding stage so it is imperative for network coding to be parallelized to reduce time delay when encoding/decoding. In this paper, we compare the performance of parallelized decoders for random linear network coding (RLC) and pipeline network coding (PNC), a recent development in order to alleviate problems of RLC. We also compare multi-threaded algorithms on multi-core CPUs and massively parallelized algorithms on GPGPU for PNC/RLC.

This work studies the potential and impact of the FRANC network coding protocol for delivering high quality Dynamic Adaptive Streaming over HTTP (DASH) in wireless networks. Although DASH aims to tailor the video quality rate based on the available throughput to the destination, it relies on the TCP protocol for reliability in data delivery. TCP is known to drop its throughput performance by several fold in the presence of even 1% or 2% packet losses, which are common in wireless systems. This will force DASH to settle at a much lower video resolution, thus reducing the user's quality of experience. We show that the use of FRANC, an adaptive network coding protocol that provides both low delay and high throughput to upper layers, as a reliability mechanism for TCP can significantly increase video quality. As part of our analysis, we benchmark the performance of various TCP versions, including CUBIC, Reno, Veno, Vegas, and Westwood+, under different packet loss rates in wireless systems using a real testbed with Raspberry Pi devices. Our goal was to choose the most promising TCP version in terms of delay performance, in this case TCP Reno, and make a fair comparison between TCP running alone and using FRANC underneath for reliability. Our demonstrator with DASH in Raspberry Pi devices using the DASH benchmark, shows that the video rate delivered is 4× higher when using FRANC. Even in harsh packet loss conditions, FRANC is able to deliver higher data rates (increase 4×), while experiencing significantly shorter (decrease 10×) video lags.