Search

This paper considers a swarm of Unmanned Aerial Vehicles (UAVs) managed by a ground station. These UAVs may experience highly varying channel gains and collisions when they transmit to the ground station. To this end, we introduce a novel Learning Medium Access Control (L-MAC) for multi-rate UAVs and equip the ground station with Successive Interference Cancellation (SIC) capability. The ground station uses L-MAC to learn a Time Division Multiple Access (TDMA) schedule/frame length that yields the highest throughput. UAVs, on the other hand, use L-MAC to learn the best transmission slot and data rate for a given frame length. Our extensive simulation results show that L-MAC achieves up to five times higher throughput as compared to the well-known Aloha protocol. Specifically, L-MAC achieves a throughput of 500 kbps as compared to 100 kbps for Aloha. In comparison, Aloha with SIC achieves a throughput of 300 kbps for the same network scenario. On the other hand, the throughput of L-MAC is as most half that of the case when the ground station has perfect channel state information. Our results also show that the frame length is always set to around 60–75% of the total number of UAVs.

A key advance in enabling higher wireless mesh network capacity is allowing routers to transmit or receive (MTR) from multiple neighbors simultaneously over the same frequency. Achieving this capacity, however, is predicated on a link scheduler that is able to capitalize on the MTR capability of nodes to activate the maximum number of active links, and also to derive the shortest schedule that ensures all links are activated at least once. To date, existing schedulers do not consider the transmission or air-time of packet(s). Henceforth, this paper fills this gap and propose to derive the shortest superframe length, defined as the end time of the last transmitting link. Our scheduler, called A-TxRx, greedily adds links whenever a link finishes its transmission. As a result, unlike previous schedulers, links can start transmitting/receiving as soon as there is no conflict. We have evaluated the performance of A-TxRx in various network configurations, and compared it against two state-of-the-art approaches: 2P and JazzyMAC. The results show A-TxRx outperforming these algorithms significantly, especially when the network becomes denser. Specifically, the superframe length of A-TxRx is typically less than half of 2P and JazzyMAC, with 60 % more concurrently transmitting links.

This letter considers routing in delay tolerant networks whereby nodes have semi-predictable mobility patterns within a time period. We propose a mobility-based routing protocol (MBRP) where nodes construct a space–time graph dynamically. As the space–time graph may be incomplete, MBRP presents a heuristic that evaluates encountered nodes based on their recorded mobility patterns in order to disseminate a finite number of bundle replicas. Simulation results, over a service quality metric comprising of delivery, delay and overhead, show that MBRP achieves up to 105 % improvement as compared to four well-known routing protocols. Finally, MBRP is capable of achieving 50 % of the performance attained by the optimal algorithm, whereby all nodes are preloaded with a space–time graph.

Information Retrieval (IR) systems aim to retrieve data that satisfies certain requirements and constitute an important service in many types of networks, including Delay/Disruption Tolerant Networks (DTNs). In current DTN based IR systems, the data that satisfies a query is assumed to be stored on a single node. Therefore, once a node receives a query in which it has the corresponding data, the query can be resolved completely. However, in scenarios where a query requires data from multiple nodes, these IR systems may fail. Henceforth, in this paper, we propose Distributed Data-Centric Information Retrieval (DDC-IR), a data centric IR system that supports all query types; e.g., continuous and complex. More importantly, it is designed specifically to operate in DTNs. It also incorporates a new packet, aka Query Reply Packet, that includes both a query and one or more replies. We show how this packet facilitates efficient query resolution and enables data centric routing. In addition, it uses caching so that nodes store popular queries that has the effect of speeding up query resolution. We have conducted an extensive simulation study to compare DDC-IR to state of the art IR systems using the popular Random Waypoint model and a trace-file containing student movements on a campus. The results show that DDC-IR is able to resolve 50 % more queries and has an 80 % lower buffer occupancy level than existing IR systems. We also tested DDC-IR in networks with varying sizes. For networks with 100 nodes, DDC-IR is able to resolve queries while current IR systems fail to resolve any queries. In particular, when the number of nodes increases, current IR systems fail to resolve any queries, whilst DDC-IR is able to resolve complex and continuous queries. The influence of the number of sub-queries on query resolution time is also studied. Specifically, when the number of sub-queries in a complex query increases from five to nine, DDC-IR uses 50 % more time to resolve a query. In comparison, prior IR systems fail to resolve any queries.

Broadcast is a fundamental operation in wireless sensor networks (WSNs). Given a source node with a packet to broadcast, the aim is to propagate the packet to all nodes in a collision free manner whilst incurring minimum latency. This problem, called minimum latency broadcast scheduling (MLBS), has been studied extensively in wireless ad-hoc networks whereby nodes remain on all the time, and has been shown to be NP-hard. However, only a few studies have addressed this problem in the context of duty-cycled WSNs. In these WSNs, nodes do not wake-up simultaneously, and hence, not all neighbors of a transmitting node will receive a broadcast packet at the same time. Unfortunately, the problem remains NP-hard and multiple transmissions may be necessary due to different wake-up times. Henceforth, this paper considers MLBS in duty cycled WSNs and presents two approximation algorithms, BS-1 and BS-2, that produce a maximum latency of at most $$(\Delta -1) TH$$ ( Δ - 1 ) T H and $$13TH$$ 13 T H respectively. Here, $$\Delta$$ Δ is the maximum degree of nodes, $$T$$ T denotes the number of time slots in a scheduling period, and $$H$$ H is the broadcast latency lower bound obtained from the shortest path algorithm. We evaluated our algorithms under different network configurations and confirmed that the latencies achieved by our algorithms are much lower than existing schemes. In particular, compared to OTAB, the best broadcast scheduling algorithm to date, the broadcast latency and transmission times achieved by BS-1 is at least $$\frac{1}{17}$$ 1 17 and $$\frac{2}{5}$$ 2 5 that of OTAB respectively.

Epidemic and ts variants are one main category of routing protocols in delay tolerant networks. They are known to be simple, have low latency and high delivery ratio. However, to date, virtually no work has investigated their suitability in delivering multicast bundles. To this end, we have conducted extensive and comprehensive simulation studies to answer this question. Specifically, we compare their delivery ratio, and buffer occupancy level in various multicast scenarios where nodes move according to both random way point model and a trace file containing records of student movements on a campus. Moreover, we tested the influence of key factors such as anti-entropy session, multicast group size, the number of groups and subscribers forwarding policies. We found that, achieving a high delivery ratio is dependent on nodes using anti-entropy session, relay/subscriber ratio and subscribers working as relays. Additionally, we also observed that relay nodes that use epidemic variants experience high buffer occupancy level. In particular, protocols such as epidemic with encounter count (EC) result in 100 % buffer utilization. To solve this problem, we propose EC quota, in which bundles can only be transmitted to $$k$$ relay nodes. Our experiments prove that EC quota is capable of reducing the buffer occupancy level of relay nodes by 15–100 % whilst maintaining high delivery ratios.

Wireless Sensor Networks (WSNs) have many characteristics that are attractive to a myriad of applications. In particular, nodes employ multi-hop communications to collaboratively forward sensed data back to one or more sinks. In this context, reducing the end-to-end delay between the sink and sensor/source nodes is of interest to many applications. In particular, those that require a fixed, upper bound on end-to-end delays. To this end, we focus on bounding the end-to-end delay from the sink to each source. We first formulate the problem as a Binary Integer Program (BIP). As the problem is NP-hard, this paper proposes and studies two centralized, heuristic algorithms: Tabu and Domino. The key approach used by both algorithms is to determine the minimal number of extra wake-up slots required by a given network in order to ensure the delay of all end-to-end paths is within a given bound. We conducted two sets of experiments. The first set compares BIP, Tabu, and Domino in WSNs with up to 80 nodes. These experiments serve to compare the proposed algorithms against BIP, which become computationally expensive in large scale WSNs. The results show that, compared to BIP, the number of additional wake-up times generated by Tabu and Domino are within 5 and 10 % of the optimal solution. In the second set of experiments, which evaluates the algorithms in WSNs with 100---500 nodes, the average number of extra wake-up slots activated by Domino is 13 % greater than Tabu. These algorithms have a time complexity of $${\mathcal {O}} (\alpha n^2 T + n^3)$$ O ( ? n 2 T + n 3 ) and $${\mathcal {O}}(n^3)$$ O ( n 3 ) respectively, where $$n$$ n is the number of nodes, $$T$$ T is the number of slots in one period, and $$\alpha$$ ? is the maximum number of iterations carried out by the Tabu algorithm.

Asynchronous duty cycle Medium Access Control (MAC) protocols do not require global synchronization because nodes determine their wake-up schedule independently. As a result, these MACs have superior performance to those that employ synchronous duty-cycles in terms of energy expenditure, and advantageously, they are simple to implement. A key limitation is that they do not support efficient broadcast. A node needs to transmit a broadcast packet multiple times via unicast because only a subset of its neighbors may be awake at any given point in time. To address this problem, this paper proposes a centralized and distributed asynchronous broadcast algorithm that achieves minimal broadcast latency and redundancy. In addition, it uses a novel asynchronous MAC protocol that ensures all neighbors of a broadcasting node are awake to receive a broadcast. The performance of our algorithms is evaluated under different network configurations. We show via extensive simulation studies that our algorithms have near optimal network performance in terms of broadcast latency. In particular, compared to OTAB, the best broadcast scheduling algorithm to date, the broadcast latency and transmission times achieved by our designs are 1/5 and 1/2 that of OTAB, respectively.

High data rate wireless technologies are now becoming a reality, and have spurred the development of new applications that were previously hindered by the lack of capacity. In particular, it is now possible to stream high definition movies from a laptop to a sound system and flat screen television seamlessly. One of the key enablers of such applications is the IEEE 802.15.3 medium access control (MAC) protocol, which is designed to support bandwidth intensive applications in wireless personal area networks (WPANs). A key observation is that a significant number of multi-media applications rely on the transmission control protocol (TCP). Unfortunately, little works have conducted a thorough performance study of TCP over the IEEE 802.15.3 MAC. Moreover, the IEEE 802.15.3 specification does not specify any strategies for allocating time slots. This paper therefore contributes to the current state-of-the-art in the following manners. From our extensive analytical and simulation studies, we reveal the impacts of different channel time allocation methods and acknowledgment policies on the performance of TCP, with particular attention on round trip time, congestion window growth, and packet recovery. We then present the following guidance to application developers: (1) channel time allocations (CTAs) should be distributed evenly over the superframe and have durations determined by TCP’s maximum congestion window, (2) CTA positioning has no impact on TCP’s performance, and (3) the specified delayed acknowledgment policy needs to be augmented with an adaptive algorithm that adjusts its burst size dynamically to varying bit error rate (BER).

In active networks (ANs) programs can be injected into routers and switches to extend the functionalities of the network. This allows programmers to enhance existing protocols and enables the rapid deployment of new protocols. The main objective of this paper is to show why ANs are ideal in solving the problem of connection rerouting and how current end-to-end based approaches can be enhanced. In this paper we propose a new model called active connection rerouting (ACR). In the ACR model, programs are dynamically injected into switches/routers in mobile networks to faciliate efficient connection rerouting during mobile host (MH) migration. We show how connection rerouting can be performed efficiently within the network. The ACR model uses a two stage optimization process: (i) path extension and (ii) lazy optimization. Unlike previous work on two stage connection rerouting ACR has the following properties: elimination of loops within switches/routers and incremental optimization which minimizes buffer requirements and maximized path reuse. ACR performs well in all topologies. Our experimental results show that ACR is efficient and scalable and it performs well in all topologies.

Active networks (ANs) are a new paradigm in computer networking. In ANs, programs can be injected into routers and switches to extend the functionalities of the network. This allows programmers to enhance existing protocols and enables the rapid deploymentof new protocols. Little work has been done in the area of multicast routing in heterogeneous environments. In this paper, we propose AMTree, an AN-based multicast tree that is bidirectional, optimizable on demand and adaptive to source migration. We show how ANs can be exploited to enable multicast tree to be modified and optimized efficiently. By filtering unnecessary signaling messages, maintaining minimal storage at routers and incorporating features from shared-tree methods we are able to achieve a scalable solution. Furthermore, we introduce an AN-based optimisation algorithm that is executed on demand by receivers. Besides that we introduce a fast rejoin protocol for receiver migration that makes no assumptions about the existence of multicast services in foreign networks. The performance of AMTree is compared to those of the bidirectional home agent (HA) method and the remote subscription method. We found that compared to the bidirectional HA method AMTree has a much lower handoff and end-to-end latency. Unlike the bidirectional HA method where end-to-end latency increases as the mobile host (MH) migrates further away from its HA, AMTree's latency remains fairly constant. We found that after optimization, the resulting tree's end-to-end latency to be comparable to the remote subscription method but without the need for building a new multicast tree after each handoff.

Broadcast is a fundamental operation in wireless networks. To this end, many past studies have studied the NP-hard, broadcast problem for always-on multi-hop networks. However, in wireless sensor networks, nodes are powered by batteries, meaning, they have finite energy. Consequently, nodes are required to have a low duty cycle, whereby they switch between active and sleep state periodically. This means that a transmission from a node may not reach all of its neighbors simultaneously. Consequently, any developed broadcast protocols must consider collisions and the wake-up times of neighboring nodes. Henceforth, this paper studies the minimum latency broadcast scheduling problem in duty cycled multi-hop wireless networks (MLBSDC), which remains NP hard. The MLBSDC problem aims to find a collision-free schedule that minimizes the time in which the last node receives a broadcast message. We propose a novel algorithm called CFBS that allows nodes in different layers of the broadcast tree to transmit simultaneously. We prove that CFBS produces a latency of at most . Here, T denotes the number of time slots in a scheduling period, and H is the optimal broadcast latency obtained from the shortest path tree algorithm assuming no collision. We also show that the total number of transmissions is at most 4(T+2) times larger than the optimal value. The results from extensive simulation show that CFBS has a better performance than OTAB, the best broadcast scheduling algorithm to date. In particular, the broadcast latency achieved by CFBS is up to that of OTAB.