Wireless Communication Systems in a Swarm of Autonomous Flying
Contents
1. AND COMMUNICATION PROTOCOLS most probably lead to the network being swamped with data packets Last but not least each of the mobile nodes must be carrying batteries onboard in order to supply themselves with power But since the power supply is limited the processing power will also be limited which will further limit the services and the application supported by the nodes 3 4 2 Comparison of the Proactive and Reactive Routing Protocols In this section the reactive and proactive protocols are to be compared by consider ing the design challenges presented in the previous section First of all on demand reactive protocols are generally more efficient than proactive ones As depicted in Table 3 1 on demand protocols minimize power consump tion since routes are only established when required Proactive protocols require periodic route updates to keep information current and consistent which leads to a worse performance They also maintain multiple routes that might never be needed adding unnecessary routing overheads Second of all proactive routing protocols provide better quality of service than on demand protocols since in the former case the available routes are constantly being updated and therefore kept up to date This minimizes at the same time the end to end delay Last but not least if we consider the throughput we notice that proactive protocols perform better than reactive protocols JG07 Reactive Protoc2. Settling y 130us 130s Bi 4 25 1 x E at y 1 j CE 0 CE 0 E AY Y Y x Y x x Y 38 y ra x E x y Pag RX 066 00 TX Mode ___ an A N 1 x i y CE A FIFO not empty Figure 2 1 Radio control state diagram NS08 Air Data Rate The nrf24L01 supports different air data rates 250kbps 1Mbps 2Mbps On one hand a lower air data rate gives a better receiver sensitivity than higher air data 2 1 TRANSCEIVER NRF24L01 9 rates On the other hand though a high air data rate gives lower average current consumption and reduced probability of on air collisions NS08 RF Channel Frequency The channel frequency can be set in a range from 2 4GHz to 2 525GHz in steps of 1MHz To set up the channels for 2Mbps data transfer rate the channel spacing between two channels should be at least 2MHz The transmitter and the receiver must use the same channel frequency in order for them to be able to exchange packets NS08 Power Amplifier In order to save current the power amplifiers amplification for transmission can be set up in four different steps Table 2 1 shows the output power according to the current consumption NS08 RF output power DE curren consumption 0dBm 11 3mA 6dBm 9 0mA 12dBm 7 5mA 18dBm 7 0mA Table 2 1 RF output power setting for the nRF24L01 NS08 2 1 2 Enhanced ShockBurst The nrf24101 transceiver has a build in protocol c
3. send by the microcontroller and defines the operation In the signal diagrams the command bits are labelled with a Cn For this byte the transceivers sends the current content of the status register labelled with Sn After that the actual data marked with Dn is also sent In contradiction to the bits which are ordered from MSBit Most Significant Bit to LSBit Leased Significant Bit the bytes are ordered from LSByte to MSByte miso AA Figure 2 7 SPI write operation NS08 2 2 LPC2103 EDUCATION BOARD 15 IRQ Pin and CE Pin There are two additional pins to control the transceiver e the CE pin to switch between active and passive mode e the IRQ pin which can show the TX_DS IRQ send successful interrupt RX_DR IRQ packet received interrupt or MAX_RT IRQ maximum number of retrans mits interrupt 2 2 LPC2103 Education Board The LPC2103 Education Board is the hardware basis of the network nodes The board is build around a microcontroller based on the ARM7TDMI chip architecture It has 32KBytes program Flash and 8KBytes SRAM EA12 2 2 1 Expansion Connector The 20 pos expansion connector is used to connect the transceiver with the micro controller Table 2 2 shows the relation between the microcontroller pins the pins of the connector and the external pins of the transceiver EA12 LPC2103 Connector nrf24101 GND pin 20 GND P0 15 EINT2 pin 14 IRQ P0 5 MISOO pin 4 MISO P0 6 MOSIO pin 5 MOSI P0 4 S
4. should be to encrypt the messages handled by the network CHAPTER 5 CONCLUSIONS AND OUTLOOK 32 33 Appendix A Content of the CD ROM edulpc2103 examples examples zip Zip file with all example codes from the Embedded Artists website 0011062103 ide lpc21isp 183 tar gz Source code of Ipc2lisp lpc2lisp is a portable command line ISP for Philips LPC2000 family and Analog Devices ADUC70xx edulpc2103 ide linux gnu chain for armelf tar bz2 64bit linux precompiled GNU Toolchain for arm elf cross compiling edulpc2103 ide win LPC2xxx gcc newlib v2_4 0 1 exe IDE for Windows XP requires GNU Toolchain edulpc2103 ide win bu 2 18_gcc 4 2 2 c c _n1 1 15 0 gi 6 7 1 exe GNU Toolchain for arm elf cross compiling for windows needed for the IDE edulpc2103 manuals data sheet lpc2103 pdf Data sheet for the LPC2103 mi crocontroller edulpc2103 manuals edu_lpc2103_baseboardSchematic_v1_1 pdf board schemat ics of the LPC2103 Education Board edulpc2103 manuals LPC2103_Education Board Users Guide pdf User guide for the LPC2103 Education Board edulpc2103 manuals manual arm7tdmi s pdf Reference manual for the ARM7tdmi s architecture edulpc2103 manuals QuickStart_Guide Version 1 1 pdf QuickStart Program Development User s Guide for the LPC2103 Education Board edulpc2103 manuals user manual lpc2103 pdf User manual of the LPC2103 microcontroller 34 APPENDIX A CONTENT OF THE CD ROM edulpc2103 rtos
5. the destination node Each node exchanges its distance estimates for all other network nodes with each of its immediate neighbors Such algorithms behave poorly leading to routing loops and slow convergence in a dynamic environment Type of Cast Property e Unicast Unicast forwarding refers to a one to one communication the sender trans mits data packets to a single destination e Geocast The data packets are delivered to all the nodes situated in a specified ge ographical area Geocasting can therefore be also seen as a form of restricted broadcasting e Multicast The sender transmits a message to multiple destinations e Broadcast Each sent message is received by all the nodes located in the transmission range of the sender a distance from one hop Sheduling e Proactive routing protocols e Reactive on demand routing protocols 3 4 PROTOCOL CHOICE 21 3 3 1 Proactive routing protocols Proactive routing protocols are table driven they attempt to maintain consistent up to date routing information between every pair of nodes in the network by proac tively propagating route updates at fixed time intervals 409 In order to store the routing information each node maintains a so called routing table Each of the nodes responds to changes in the network topology by propagating updates throughout the MANET in order to maintain a consistent network view RT99 3 3 2 Reactive on demand routing protocol
6. this purpose the node looking for a route starts a route re quest by broadcasting a RREQ message This RREQ message has the following structure e first byte message ID RREQ e second byte target node e third byte node ID e following bytes Node IDs of all nodes that have been passed forming a route Vector If a route request reaches a node it is checked whether the node is already include in the RREQ message If so the message is discarded which ensures that the RREQ message won t get stuck in a routing loop If not the node ID is added to RREQ message and the message is broadcasted again At some point one 25 Figure 4 1 Hello Message 4 3 ROUTE REQUEST 26 CHAPTER 4 PROTOCOL or multiple versions of this RREQ message will reach the target node There the routes are stored in the routing table The route replay message is built and sent back along the route the related route request message took These RREP messages have the following structure e first byte message ID RREP e following bytes node IDs in order of the route from the requesting node to the target node On the way back to the RREQ source every node safes the route in both directions in his own routing table Every node along the route will therefore know in which direction he has to forward a message along the route Since the RREQ message is broadcast through the hole network it is h
7. CKO pin 3 SCK P0 7 SSELO pin 6 CSN P0 9 GPIO pin 8 CE VDD pin 19 VCC Table 2 2 The connection between LPC2103 and nrf24101 EA12 NXPO9 The communication between microcontroller and transceiver is described in Section DAS 2 2 2 UART Port UARTO on the LPC2103 is connected to a USB to serial bridge chip FT232RL from FTDI The serial interface is not a full interface only the receive and trans mit signals are connected to UARTO This allows to connect the microcontroller to a computer to program the microcontroller and print messages from the microcon troller on a terminal EA12 CHAPTER 2 HARDWARE 16 17 Chapter 3 Intoduction to MANET and Communication Protocols Kleinrock Kle03 describes ad hoc networking technology in one of his papers as 8 blend of nomadicity embeddedness and ubiquity In a network of the future users and computing devices will be able to connect to such a network conveniently and even transparently Computing and communication capabilities will not only be restricted to standard electronic devices but every gadget can afford to embed a considerable amount of intelligence On a global basis devices in the network will be able to rely on other devices to relay packets for them if necessary CLL This way the world will be heterogeneously networked by a vast invisible global infrastructure Kle03 considers Kleinrock Since t
8. L Kleinrock An internet vision the invisible global infrastructure Ad Hoc Networks 2003 Nordic Semiconductor Otto Nielsens vei 12 7004 Trondheim nRF24L01 Single Chip 2 4GHz Transceiver Product Specification v1 0 2008 http www nordicsemi com NXP LPC2101 02 03 User manual 2009 http www nxp com Elizabeth Royer and Chai Keong Toh A review of current routing proto cols for ad hoc mobile wireless networks JEEE Personal Communications 1999 409 CLL EA12 JG07 Kle03 NSO8 NXP09 RT99
9. RTOS_LPC2xxx_v2 2 zip Realtime operating system from Embedded Artists nrf24101 manual data sheet nRF24L01P pdf Data sheet for the nrf24101 transceiver nrf24101 tutorials nrf24101_tutorial_O pdf Tutorial for the nrf24101 transceiver explaining the transceiver mrf24101 tutorials nrf24101_ tutorial 1 zip Tutorial for the nrf24101 transceiver with a simple demo for an ad hoc connection mrf24101 tutorials nrf24101 tutorial 2 zip Tutorial for the nrf24101 transceiver with details on the Enhanced ShockBurst protocol mrf24101 tutorials nrf24101 tutorial 3 zip Tutorial for the nrf24101 transceiver with details on multi pipelining mrf24101 tutorials nrf24101 tutorial 4 zip Tutorial for the nrf24101 transceiver with details on Cryptography with ARC4 report report pdf This report as a pdf file report report zip The Latex source files for this report report templates pp report zip Template use for the report slides 07 11 2011 zip Slides from the first presentation slides 12 12 2011 zip Slides from the secound presentation slides 20 01 2012 zip Slides from the third presentation slides slidetemplate_Isr_2011 10 05 zip Template for the sildes src src zip Soure code of the programs for the microcontroller src refman pdf Dokumentation for the soure code LIST OF FIGURES List of Figures 1 1 Swarm of Flying Helicopters sa sa sanieren 2 1 Radio control state diagram NS08 2 2 2 Enhan
10. WIRELESS COMMUNICATION SYSTEMS IN A SWARM OF AUTONOMOUS FLYING HELICOPTERS eingereichte STUDIENARBEIT von Ludwig John B Sc Alexandra Marinescu geb am 21 08 1986 geb am 28 08 1989 wohnhaft in wohnhaft in Billrothstr 18 Kreuzhofstr 6 81369 Munchen 81476 Munchen Tel 089 51665558 Tel 0176 83260686 Lehrstuhl fiir STEUERUNGS und REGELUNGSTECHNIK Technische Universitat Munchen Univ Prof Dr Ing Univ Tokio Martin Buss Univ Prof Dr Ing Sandra Hirche Betreuer Prof Dr sc nat Jorg Conradt amp M Sc Cristian Axenie Beginn 07 11 2011 Zwischenbericht 12 12 2011 Abgabe 20 01 2012 Abstract The ability to communicate and collaborate is essential for any multi agent tasks WLAN is a simple possibility to integrate such communication into existing net works Since the various commercial modules for mobile robots are typically expen sive operate on relatively low transmission speed and also don t have robot agent specific dimensions the aim of this project is to design and implement a task specific communication protocol aimed at offering point to point communication in networks with constantly and rapidly changing topologies Zusammenfassung In der Lage zu sein zu kommunizieren und zusammen zu arbeiten ist essenziell fir jedes Multiagenten System WLAN ist die einfachste Losung um Kommu nikationm glichkeiten in ein existierendes Netzwerk zu integrieren Da die ver schieden kommerziel verf gbaren Module f
11. alled Enhanced ShockBurst Enhanced ShockBurst is a packet based data link layer that features automatic packet assembly and timing automatic acknowledgement and retransmissions of packets NS08 The build in protocol allows to easily set up a connection between two transceivers and reduces the load on the microcontroller since all protocol operation are per formed on the transceiver Packet Format Figure 2 2 shows the structure of an Enhanced ShockBurst packet e Preamble The preamble is a one byte long alternating sequence of 1 and 0 aimed at synchronizing the receivers demodulator to the incoming bit stream 10 CHAPTER 2 HARDWARE e Address The address ist used to logically distinguish between the different pipelines in the same RF channel According to the needs of the application the address can have a length of three four or five bytes e Packet Control Field The packet control field as showed in Figure 2 3 contains control infor mation for the packet The payload length information is mandatory to allow dynamic payload length The PID is used to detect packets that were retransmitted because the acknowledgement was lost The NO_ACK flag allowes to disable the acknowledgement for certain packets e Payload The payload is the user defined content of the packet It can also have a variable length which can be dynamically changed e CRC The CRC is the mandatory error d
12. ansmitting packets and the RX mode for receiving pack ets In addition to that there are three passive modes for power saving The power down mode has the lowest power consumption In this mode only the register values are maintained and the SPI is kept active In standby I mode the crystal oscillator is partly active which consumes more current but allows faster switching to one of the active modes The last passive mode is standby II Here the nRF24L01 waits CHAPTER 2 HARDWARE for packets uploaded to the TX FIFO register As soon as a new packet arrives the packet can be send without the 130s delay for the Phase locked loop PLL set tling The interaction between these states is shown in the state diagram in Figure 2 1 NS08 Legend Undefined i i Transision state gt Recomended Vcc gt 1 9V Undefined operating mode Possible operating mode NS Recomended Power transision i S rn Possible Sa we er Power __ 2 transision gag OE Down Sig eee Y T 1 Start Up N 1 1 1 1 x 2 7 7 1 PW_UP 0 y 1 1 1 4 i FIFO empty 1 1 ann CE 1 1 i PRIM_RX 1 PRIM_RX 0 i CE 1 gt k 1 i 1 A FIFO not empty x 1 CE 1 ES 4 0 PW_UP 0 PRIM 0 1 I 1 cae 1 gt hg e 3 a 72 a PW_UP 0 1 1 n 1 RX Settling
13. can be delivered in a timely manner Also the well known routing protocols used in wired networks cannot be used for mobile ad hoc networks since they are not considering the main property of ad hoc networks node mobility 3 3 Taxonomy of MANETs JG07 In order to classify and evaluate the different MANET protocols several criteria reflecting fundamental design and implementation choices have to be considered Communication Model e Multi channel communiation Multi channel protocols are low level routing protocols which combine chan nel assignment and routing functionality e Single channel communiation Single channel protocols relay on specific link layer behaviours and are Carrier Sense Multiple Access Collision Avoidance CSMA CA oriented Structure e Uniform protocols None of the nodes takes a distinguishing role in the routing scheme There is no hierarchical structure in the network e Non uniform protocols Routing complexity is being limited by reducing the number of nodes partic ipating in the routing computation 20 CHAPTER 3 INTODUCTION MANET AND COMMUNICATION PROTOCOLS State Information e Topology based protocols The topology based protocols rely on link state protocols The participat ing nodes maintain large scale topology information Each node makes decisions based on complete topology information e Destination based protocols Each node maintains a distance and vector next hop to
14. ced ShockBurst packet format NS08 2 3 Control field NS08 o rs ele RER 2 4 PTX operations in Enhanced ShockBurst NSO8 2 5 PRX operations in Enhanced ShockBurst NS08 2 6 SPI read operation NS08 eto u e ita e 2 7 SPI write operation NS08I Sisa a ae 3 1 MANET Topolgy with One Possible Route AL Hello Message LE AI E ae 8 4 2 Route Request and Route Reply Functions 4 3 Route Request and Route Reply Functions with Fallback Routes 4 4 Message Send Function 4000 pace ode oe Ree Oe ek LIST OF FIGURES 36 LIST OF TABLES 37 List of Tables 2 1 RF output power setting for the nRF24L01 NSO8 9 2 2 The connection between LPC2103 and nrf24101 EA12 NXPO9 15 3 1 Comparison of the Proactive and Reactive Routing Protocols 22 LIST OF TABLES 38 BIBLIOGRAPHY 39 Bibliography Algorithms and Protocols for Wireless and Mobile Ad Hoc Networks John Wiley and Sons 2009 Yuanzhu Chen Arthus Liestman and Jiangchuan Liu Clustering algo rithms for ad hoc wireless networks Embedded Artists AB Ole Roemers vaeg 12 Ideon Technology Park SE 223 70 Lund LPC2103 Education Board Users Guide 2012 http www embeddedartists com Geetha Jayakumar and Ganapathy Gopinath Ad hoc mobile wireless networks routing protocols a review Journal of Computer Science 3 2007
15. duced as the first byte of every message This way the communication system can process the incoming message in an appropiate way depending on the label that was assigned 24 CHAPTER 4 PROTOCOL 4 2 Neighbour Search The biggest challenge in designing a MANET protocol is probably the constant and unpredictable changing of the network topology Recognizing and being able to communicate immediately with the neighbours in transmission range of one node one hop distance is therefore essential for a working communication system As depicted in Figure 4 1 the individual nodes broadcast in short intervals with a predefined length a Hello message in their close proximity The Hello message has the following structure e first byte message ID HELLO e second byte node ID of the source In responde to the received Hello message the nodes in the transmission range of the source node send back to the address in the second byte a Hello Reply message which has the following structure e first byte message ID HELLOREP e second byte node ID of the source After decoding the message ID the source node updates the entries of its routing table by adding a direct route to the neighbours whose addresses are consistent with the second byte of the received Hello Reply 4 3 Route Request To get informations about routes beyond the direct neighbours a route exploration has to be initiated For
16. e No Packet in No Yes No TX FIFO No gt Packet in e TX Settling and TX FIFO ES Packet assembly Packet in TX FIFO Yes Yes Auto Re No Set TX_DS IRQ No Transmit enabled RX Mode and Packet Disassembly Standby Mode II No No ARD elapsed ACK with payload of retries ARC Put payload in RX FIFO Set TX_DS IRQ Set RX_DR IRQ 12 Set MAX_RT IRQ TX mode retransmit jacket last p Figure 2 4 PTX operations in Enhanced ShockBurst NS08 13 2 1 TRANSCEIVER NRF24L01 Start Primary RX Standby mode RX Mode Packet received No gt Put payload in RX FIFO set RX_DR IRQ Yes Auto ACK enabled Put payload in Yes RX FIFO set RX_DR IRQ Is received Packet a new packet Discard packet No Has ACK Payload Set TX_DS IRQ payload in TX FIFO TX Mode Transmit ACK TX Settling TX Mode Transmit ACK with payload Yes TX Settling Figure 2 5 PRX operations in Enhanced ShockBurst NS08 14 CHAPTER 2 HARDWARE Figure 2 6 SPI read operation NS08 There is always one byte transmitted in both directions The transmission clock SCK is defined by the master device Every read and write command for the different registers of the transceiver is encoded with a unique command word This is the first byte
17. etection mechanism in the packet It is either 1 or 2 bytes and is calculated over the address Packet Control Field and Payload Preamble Address Packet Control Field Payload CRC 1 byte 3 5 byte 9 bit O 32 byte 1 2 byte Figure 2 2 Enhanced ShockBurst packet format NS08 Payload length 6 bit PID 2 bit NO_ACK 1 bit Figure 2 3 Control field NS08 2 1 TRANSCEIVER NRF24L01 11 PTX Operation Figure 2 4 shows the flowchart of the PTX operation of a nrf24101 starting form Standby I mode If CE is set to 1 the transceiver changes to active mode and starts transmitting the payload stored in the TX FIFO If no data is present in the TX FIFO the transceiver waits in Standby I mode as long as CE is set to 1 The payload from the TX FIFO is assembled to a packet and transmitted If the auto retransmit feature is enabled and the NO_ACK flag is set to 0 the transceiver changes to RX mode to wait for the acknowledgement If no acknowledgement is received after the maximum number of retransmits the MAX_RT IRQ is set and the transceiver changes to Standby I mode In case of a successful received acknowledgement the TX_DS IRQ is set PRX Operation The flowchart in Figure 2 5 shows the PRX behavior after entering Standby I mode Packets can only be received if the transceiver is in active mode by setting CE to 1 If the auto acknowledgement feature is disabled packets are stored to RX FIFO as long as t
18. he latter is not full If the auto acknowledgement feature is enabled it first has to be checked whether a packet is new or retransmitted Only new packages are being added to the RX FIFO If the acknowledgement packet has a payload attached to it the latter is stored in the TX FIFO and the TX_DS IRQ is set If the packet requires sending back an acknowledgement the transceiver switches to TX mode long enough to send the ACK packet and returns immediately to RX mode afterwards MultiCeiver MultiCeiver is a feature that allows the nrf24101 to decode packets in RX mode for six parallel data pipes with unique addresses A data pipe is a logical channel in the physical RF channel The data pipes can be configured for individual behaviour 2 1 3 Data and Control Interface The nrf24101 transceiver is accessible via Serial Peripheral Interface SPI SPI devices communicate in master slave mode For this application the microcon troller operates as master and can therefore initiate the communication while the transceiver operates as slave Figures 2 6 and 2 7 show a typical read and write operation between microcon troller and transceiver The CSN signal is set to 0 to initiate the data frame The SPI communication runs in full duplex mode over two different connections the MISO Master Input Slave Output connection and the MOSI Master Output Slave Input connection CHAPTER 2 HARDWARE Star Primary TX Standby Mod
19. he mentioned mobile ad hoc networks MANETs have become increasingly popular over the last few years this chapter starts by giving a quick overview on their structure and main characteristics After understanding the core concept of the MANETs the different possible topologies of such networks will be introduced together with the common design constraints which have to be considered when building such a structure 3 1 Mobile Ad Hoc Networks A mobile ad hoc network is a wireless mobile network formed spontaneously without the aid of centralized administration or standard support services regularly available as on conventional networks Communication in such a decentralised network typ ically involves temporary multi hop relays with the nodes using each other as the relay routers without any fixed infrastructure They are allowed to move randomly and organize themselves arbitrarily the network s topology might therefore change rapidly and unpredictably Figure 3 1 JG07 CHAPTER 3 INTODUCTION MANET AND COMMUNICATION PROTOCOLS Figure 3 1 MANET Topolgy with One Possible Route 18 3 2 MANET PROTOCOL 19 3 2 MANET Protocol To facilitate communication within the network a routing protocol is to be imple mented in order to discover the routes between nodes and therefore be able to send messages The goal of the routing protocol is to have an efficient route establish ment between a pair of nodes so that messages
20. ighly likely that the request will lead to multiple results The additional routes are not discarded they are also saved in the corresponding routing table so that in the case that the preferred route breaks they can be used to safe the time and bandwidth it would take for a new request Figure 4 3 4 4 Message Handling In the previous part we described the packets that are needed to maintain the network structure The following section deals with the actual user specified message that is going to be fed as an input to the network 4 4 1 Packet History In order to keep track of the already sent messages that are still waiting for a route or an acknowledgement packet the node stores the former in a so called packet history The packets are stored until the successful transmission is confirmed or it ultimately fails This is the case when no route is found or the maximum number of retransmits is reached 4 4 2 Message Transmission Messages are transmitted within the payload of the MSG packet This packet has the following structure e first byte message ID MSG e second byte ID of the source node e third byte ID of the target node e fourth byte ID of this packet 27 4 4 MESSAGE HANDLING Figure 4 2 Route Request and Route Reply Functions CHAPTER 4 PROTOCOL x k 0 e k N 0 Fi 0 r O N 1 1 1 A 1 0 s Figure 4 3 Route Request and Route Re
21. k with dynamically changing topology will be performed Finally an appropriate communication protocol will be chosen followed by an overview of its actual implementation and functionality CHAPTER 1 INTRODUCTION f r i Figure 1 1 Swarm of Flying Helicopters Chapter 2 Hardware In this chapter the main hardware components of an elementary unit will be in troduced As a test environment for the MANET protocol several nodes will be communicating with each other over radio frequency RF Each of these units con sist of a transceiver as described in Section 2 1 and a microcontroller board as described in Section 2 2 2 1 Transceiver nRF24L01 The transceiver used in this project is the nRF24L01 single chip transceiver from Nordic Semiconductor It uses the 2 4 GHz ISM Industrial Scientific and Medical band All frequency bands that can be used for high frequency devices in Industrial Scientific or Medical applications are called ISM bands Furthermore the transceiver can transmit with on air data rates up to 2 Mbps and has a build in protocol called Enhanced ShockBurst The transceiver is designed for ultra low power operation NS08 2 1 1 Radio Control This section describes the radio control options and the operational modes of the nRF24L01 radio transceiver Operational Modes The nRF24101 can run in different operational modes There are the two active modes the T X mode for tr
22. ol Proactive protocol Feature Routing Load Power consumption End to End Delay Throughput Table 3 1 Comparison of the Proactive and Reactive Routing Protocols By keeping in mind both the aim of the project and the hardware constraints we evaluated the existing communication protocols and settled for the reactive on demand version which we will discuss in detail in the next chapter 23 Chapter 4 Protocol This chapter will give an overview of both the implementation of the chosen MANET protocol and its functionality As already mentioned in the previous chapter we settled for a reactive on demand protocol since we are dealing with low MCU memory SRAM Again the source initiated on demand routing generates routes only when and if the source node requests it When the latter tries sending a package to a destina tion node which either has no entry in the routing table of the source node node unknown or has a route attached which now is unavailable it initiates a route dis covery process within the network The available routes to a specific destination are all saved to the corresponding entry of the implemented routing table 4 1 Message Types The implemented communication system handles three different types of messages e Neighbour Search e Route Request e Message Handling In order to distinguish between the different types a so called message ID has been intro
23. ow well the results of this project cover the tasks described in the problem formulation and an outlook is presented on what can be done to further improve the existing communication system We managed to implement a MANET protocol which suits both the task descrip tion and the existing hardware constraints The latter resulted though in strong limitations of the payload length the routing table or even the number of saved fallback routines As a further improvemet we would therefore suggest the use of a microcontroller MCU that disposes of a bigger SRAM memory The LPC21XX family offers a broad variety of MCUs so the code could easily be adapted with only a few or even no changes at all In the short time of this practical course we were able not only to design and imple ment a tailor made MANET protocol but also to verify its basic functionality by implementing a demo application Since this takes place on the data layer we also suggest the development of an application layer aimed at allowing and promoting the external use of the implemented protocol We also consider the security of the network of great importance since the former is one of the main design challenges in building communicating and collaborating systens For the use of the proposed communication system in a real environment and under normal working conditions security issues should be considered Since wireless networks are very vulnerable to attacks part of a future work
24. ply Functions with Fallback Routes 28 4 4 MESSAGE HANDLING 29 e fifth byte Lifetime of this packet e following bytes payload of the packet The message packet is send along a route from node to node If it reaches the target node an acknowledgement message is generated and send back to the source node This ACK message looks as follows e first byte message ID ACK e second byte ID of the source node e third byte ID of the target node e fourth byte ID of this packet that should be acknowledged e fifth byte Lifetime of this packet If at any point of the route the message can t be forwarded a negative acknowl edgement is generated and sent back to the source node of the data packet This NOACK message looks as follows e first byte message ID NOACK e second byte ID of the source node e third byte ID of the target node e fourth byte ID of this packet that should be acknowledged e fifth byte Lifetime of this packet In the case of the MSG ACK or NOACK packets it can happen that due to routing errors they are send in circles In order to avoid this the number of hops are counted and stored in the fifth byte If the number exceed a certain predefined amount of hops the packets are deleted CHAPTER 4 PROTOCOL Figure 4 4 Message Send Function 30 31 Chapter 5 Conclusions and Outlook In this chapter a quick review is given as to h
25. r mobile Roboter sehr teuer sind und relativ lange bertragungszeiten haben und zus tzlich nicht die Roboter spezifis chen Abmessungen haben ist das Ziel dieses Projekts das Design und die Imple mentierung eines Aufgaben spezifischen Kommunikationsprotokolls welches daf r ausgelegt sein soll Punkt zu Punkt Kommunikation in einem Netzwerk mit sich andauernd und schnell wandelnder Topologie zu erm glichen 23 23 24 24 26 26 26 31 33 35 CONTENTS Contents 1 Introduction 2 Hardware 2 1 Transceiver nRREZALDI E local an sr e id en ee 2 131 Radio Controls 2 keh sg ah he a ah ga er 2 1 2 Enhanced Shock Bursts amp Ot Sch a ach a a he 2 A 2 1 3 Data and Control Interface 2 2 LP 2103 Education Board 2 2 sun Smart A i 2 2 1 Expansion Connector se alle a he aha dente 2 22 WARE Port BE ee 3 Intoduction to MANET and Communication Protocols 3 1 Mobile Ad Hoc Networks rs ar ar tt Lat 3 22 MANET 0 000 lec Drang 3 3 Taxonomy of MANETs 16307 0000 3 3 1 Proactive routing protocols year eb ae ab ws 3 3 2 Reactive on demand routing protocols Bt PROLOCOMC 6 taa ara an a AA des 3 4 1 Design Challenger a dws a 0 5 ach a se 3 4 2 Comparison of the Proactive and Reactive Routing Protocols 4 Protocol dl Message Types a a Be pa hohe a eS 4 2 Neighbour Search u a a
26. s The source initiated on demand routing generates routes only when and if the source node requests it When the latter tries sending a package to a destination node which either has no entry in the routing table of the source node node unknown or has a route attached which now is unavailable it initiates a route discovery process within the network This process is completed once a route is found or all possible route permutations have been examined Once a route has been established it is maintained in the corresponding routing tables along the path until the destination is no longer acces sible or the route expires 3 4 Protocol Choice In order to find the type of protocol that best fits our task description and to moti vate our final choice the general design challenges are first to be considered followed by a comparison of the reactive and proactive protocols 3 4 1 Design Challenges Probably the biggest design challenge in designing a MANET protocol is the lack of infrastructure The network does not depend on any established structure so each node acts as an independent router The constant and unpredictable changing of the network topology results in routes becoming unavailable quite frequently and packets being lost Also as already mentioned since no default router is available every single node in the network will act as a router it will forward all incoming packets which will 22 CHAPTER 3 INTODUCTION TO MANET
27. sn a A A eS eee 43 Route Request N a ee eb 4 4 Message Handling 2 3 2a ce Ba AA nahe 4 4 1 Packet History et oS Bat o os e i 4 4 2 Message Transmission e ble Da ae be wine a 5 Conclusions and Outlook A Content of the CD ROM List of Figures 4 CONTENTS List of Tables 37 Bibliography 39 Chapter 1 Introduction In a disaster recovery environment for example with no preexisting communication infrastructure the actual conditions do not allow humans to examine the disaster area on their own as it might put their lives in danger Instead a swarm of flying he licopters agents as depicted in Figure 1 1 can be used for the recognition mission Without a proper communication between them the agents will not be able to orga nize themselves into a swarm or to optimize the accomplishment of the task making the mission goals difficult to achieve Establishing communication between the individual units is therefore essential not only for this particular case but for the collective tasks in general The aim of this project is to develop a tiny printed circuit board PCB consisting of a transceiver chip and a microcontroller aimed at representing a complete stand alone communi cation system In the first part of this report the main hardware components of such an elementary chip will be introduced Then an evaluation of different communication schemes for a possibly large number of mobile nodes in a networ
Download Pdf Manuals
Related Search