Data Link Layer (Sicherungsschicht) Medium Access Control (Zugriffsverfahren)

Similar documents
EITF25 Internet Techniques and Applications L4: Network Access. Stefan Höst

Local Area Networks (LANs) SMU CSE 5344 /

Topics. Link Layer Services (more) Link Layer Services LECTURE 5 MULTIPLE ACCESS AND LOCAL AREA NETWORKS. flow control: error detection:

The Medium Access Control Scheme (MAC Layer) Reference: Andrew S. Tanenbaum, Computer Networks, 3rd Edition, Prentice Hall, 1996.

CSE 461: Multiple Access Networks. This Lecture

Computer Network Fundamentals Spring Week 3 MAC Layer Andreas Terzis

Goals. Fundamentals of Network Media. More topics. Topics. Multiple access communication. Multiple access solutions

ECE 4450:427/527 - Computer Networks Spring 2017

Link Layer and LANs 안상현서울시립대학교컴퓨터 통계학과.

CS 43: Computer Networks. 27: Media Access Contd. December 3, 2018

Multiple Access Protocols

CHAPTER 7 MAC LAYER PROTOCOLS. Dr. Bhargavi Goswami Associate Professor & Head Department of Computer Science Garden City College

Lecture 9: Bridging. CSE 123: Computer Networks Alex C. Snoeren

CS 43: Computer Networks Media Access. Kevin Webb Swarthmore College November 30, 2017

Data Link Layer: Collisions

CSE 461: Multiple Access. Homework: Chapter 2, problems 1, 8, 12, 18, 23, 24, 35, 43, 46, and 58

The Link Layer and LANs. Chapter 6: Link layer and LANs

Chapter 6 Medium Access Control Protocols and Local Area Networks

Multiple Access (1) Required reading: Garcia 6.1, 6.2.1, CSE 3213, Fall 2010 Instructor: N. Vlajic

Aloha and slotted aloha

Multiple Access Links and Protocols

Computer Networks Medium Access Control. Mostafa Salehi Fall 2008

Medium Access Control. IEEE , Token Rings. CSMA/CD in WLANs? Ethernet MAC Algorithm. MACA Solution for Hidden Terminal Problem

Physical Layer. Medium Access Links and Protocols. Point-to-Point protocols. Modems: Signaling. Modems Signaling. Srinidhi Varadarajan

Data Link Layer Technologies

ECE 333: Introduction to Communication Networks Fall Lecture 19: Medium Access Control VII

Frames, Packets and Topologies

2.1 CHANNEL ALLOCATION 2.2 MULTIPLE ACCESS PROTOCOLS Collision Free Protocols 2.3 FDDI 2.4 DATA LINK LAYER DESIGN ISSUES 2.5 FRAMING & STUFFING

IEEE , Token Rings. 10/11/06 CS/ECE UIUC, Fall

Data and Computer Communications

COMPUTER NETWORKS - Local area networks

Local area networks. Copyright

Random Access. 1. Aloha. 2. Slotted Aloha 3. CSMA 4. CSMA/CD

Multiple Access Communications. EEE 538, WEEK 11 Dr. Nail Akar Bilkent University Electrical and Electronics Engineering Department

Outline. Introduction to Networked Embedded Systems - Embedded systems Networked embedded systems Embedded Internet - Network properties

Redes de Computadores. Medium Access Control

LANs Local Area Networks LANs provide an efficient network solution : To support a large number of stations Over moderately high speed

Chapter 8 LAN Topologies

COMP476 Networked Computer Systems. Polling. Sharing the Wire. LAN Technologies. COMP476 Networked Computer Systems 1

Data Link Layer -2- Network Access

CSE/EE 461 Wireless and Contention-Free Protocols

Chapter 5 Link Layer and LANs

Lecture 8 The Data Link Layer part I. Antonio Cianfrani DIET Department Networking Group netlab.uniroma1.it

Packet multiple access and the Aloha protocol

COMMUNICATION NETWORKS NETW 501

Data Link Layer, Part 5. Medium Access Control

Chapter 12 Multiple Access 12.1

Mobile Communications Chapter 3 : Media Access

Data Link Layer -2- Network Access

ECSE-4670: Computer Communication Networks (CCN) Informal Quiz 3

Lecture 4: Wireless MAC Overview. Hung-Yu Wei National Taiwan University

CCM 4300 Lecture 5 Computer Networks, Wireless and Mobile Communications. Dr Shahedur Rahman. Room: T115

Lecture 6 The Data Link Layer. Antonio Cianfrani DIET Department Networking Group netlab.uniroma1.it

Media Access Control. Networked Systems (H) Lecture 5

CS 455/555 Intro to Networks and Communications. Link Layer

Protocols for Multiaccess Networks

COS 140: Foundations of Computer Science

Intelligent Transportation Systems. Medium Access Control. Prof. Dr. Thomas Strang

Chapter 4. The Medium Access Control Sublayer. Points and Questions to Consider. Multiple Access Protocols. The Channel Allocation Problem.

CARRIER SENSE MULTIPLE ACCESS (CSMA):

LAN. CS 4/55231 Internet Engineering. Kent State University Dept. of Computer Science

CSC 4900 Computer Networks: The Link Layer

Real-Time (Paradigms) (47)

Wireless Communications

Lecture 5 The Data Link Layer. Antonio Cianfrani DIET Department Networking Group netlab.uniroma1.it

LANs. Local Area Networks. via the Media Access Control (MAC) SubLayer. Networks: Local Area Networks

Module 5. Broadcast Communication Networks. Version 2 CSE IIT, Kharagpur

Communication and Networks. Problems

Lecture 4b. Local Area Networks and Bridges

Chapter 6 Medium Access Control Protocols and Local Area Networks

CMPE 150/L : Introduction to Computer Networks. Chen Qian Computer Engineering UCSC Baskin Engineering Lecture 16

This formula shows that partitioning the network decreases the total traffic if 1 N R (1 + p) < N R p < 1, i.e., if not all the packets have to go

ECEN 5032 Data Networks Medium Access Control Sublayer

1999, Scott F. Midkiff

1-1. Switching Networks (Fall 2010) EE 586 Communication and. November 8, Lecture 30

Chapter 4. The Medium Access Control Sublayer

Data Link Layer: Multi Access Protocols

CS 716: Introduction to communication networks. - 9 th class; 19 th Aug Instructor: Sridhar Iyer IIT Bombay

Link Layer and Ethernet

Power Laws in ALOHA Systems

Lecture 6. Data Link Layer (cont d) Data Link Layer 1-1

Link layer, LANs: outline. Chapter 5-1 Link Layer. Link layer: introduction. Link layer services

DQDB. Distributed Queue Dual Bus (DQDB) DQDB is a MAN. Unlike FDDI, DQDB is an IEEE standard: 802.6

COMPUTER NETWORKS MODEL QUESTION PAPER WITH SOLUTION. (c) Peer-to-peer processes are processes on two or more devices communicating at a

Access Technologies! Fabio Martignon

CCM 4300 Lecture 6 Computer Networks, Wireless and Mobile Communications. Dr Shahedur Rahman. Room: T115

Data Link Layer: Overview, operations

ECE453 Introduction to Computer Networks. Broadcast vs. PPP. Delay. Lecture 7 Multiple Access Control (I)

RMIT University. Data Communication and Net-Centric Computing COSC 1111/2061/1110. Lecture 8. Medium Access Control Methods & LAN

Problem Set Name the 7 OSI layers and give the corresponding functionalities for each layer.

Systems. Roland Kammerer. 10. November Institute of Computer Engineering Vienna University of Technology. Communication Protocols for Embedded

Lecture 6: Example LAN: Ethernet

Chapter 1 Basic concepts of wireless data networks (cont d)

The MAC layer in wireless networks

CSE 461 Multiple Access. David Wetherall

MULTIPLE ACCESS PROTOCOLS 2. 1

CS 4453 Computer Networks Winter

Link Layer and Ethernet

COMPUTER NETWORK Model Test Paper

COMPUTER NETWORKS UNIT 3

Transcription:

Data Link Layer (Sicherungsschicht) Medium Access Control (Zugriffsverfahren) f References: [Bossert99] [Wa0102] www.comnets.uni-bremen.de MAC - 1

LOGICAL LINK CONTROL MEDIUM ACCESS CONTROL PHYSICAL SIGNALING DATA LINK LAYER PHYSICAL LAYER ACCESS UNIT INTERFACE PHYSICAL MEDIA ATTACHMENT MEDIUM OSI-model for local networks with partition of the link layer in media access control and logical link control, as defined by IEEE 802. www.comnets.uni-bremen.de MAC - 2

Multiple Access Protocols (Vielfachzugriffsprotokolle) MAC: Medium (Multiple) Access Control Influence Medium (Übertragungsmedium) Network Topology (Netztopologie) Performance Aspects www.comnets.uni-bremen.de MAC - 3

Multiple Access Protocols Random Access Methods Pure Aloha Slotted Aloha Carrier Sensing Reservation Aloha Scheduling Methods decentralized: Token Passing Register Insertion Ring DQDB Reservation Methods TDMA Time Division Multiple Access FDMA Frequency Division Diii Multiple Access CDMA Code Division Multiple Access Aloha with collision resolution methods centralized: Polling www.comnets.uni-bremen.de MAC - 4

Competitive Methods Random Access Schemes (Zufallszugriff) Every station with data to send can principally access the channel Transmission successful, when only one station accesses the medium Concurrent access of more than one station at the same time leads to errors www.comnets.uni-bremen.de MAC - 5

Scheduling Methods (Zuteilungsverfahren) Perfectly scheduled Centralized: Polling: a master station polls each station in turn Decentralized: all stations are equal (Token Passing, Register Insertion, DQDB) www.comnets.uni-bremen.de MAC - 6

Reservation Methods (Reservierungsverfahren) Examples: TDMA, FDMA, CDMA Management of a channel by a central station Channel with synchronous multiplexing techniques with subchannels Assignment of one or more channels on request Request in a fixed subchannel through random access or scheduling method Combined Methods www.comnets.uni-bremen.de MAC - 7

Network Topology Polling: (usually) star topology master in the center Ring / Bus: pairs of fiber channels, often separate rings or busses in both directions (dual ring, dual bus) Mobile Radio Networks usually use random access methods for reservation Reliability (Ausfallsicherheit) and Robustness: Polling? Critical: error in central station Random access methods? Critical: high load www.comnets.uni-bremen.de MAC - 8

Performance Evaluation of Methods Offered Traffic (Verkehrsangebot) Transmission Time (Übertragungszeit) Throughput (Durchsatz) Which methods are to be preferred for high and low traffic? How does the offered traffic influence throughput and transmission time? Random access methods lead to lower transmission i times for low or high h traffic? low Reservation and Scheduling Strategies lead to lower transmission times for low or high traffic? high www.comnets.uni-bremen.de MAC - 9

Aloha Protocols Development at the University of Hawaii end of the 60s for the transmission of packet data in a radio network between terminals and a central computer Basic method of all random access methods Easy to implement, no complex coordination Unsuccessful transmissions are repeated Robust against transmission errors Used in mobile radio networks, e.g., GSM (Global System for Mobile Communication) or TETRA (Trans-European Trunked Radio, digitaler Bündelfunk) for reservation Collisions www.comnets.uni-bremen.de MAC - 10

Assumptions for the Evaluation of Collision Resolution Methods (Kollisionsauflösungsverfahren) (1) m sending stations and exactly one receiver station One jointly used channel from all sending stations to the receiver Channel close to error free Data packets generated in all stations with Poisson distribution with rate m, all packets of equal length, X transmission time of a packet Immediate Transmission for Pure (Unslotted) Aloha Transmission in the next time slot for Slotted Aloha Transmission successful when only one sender accesses the channel during the transmission Collisions for concurrent access Immediate feedback: sender knows immediately whether or not the transmission was successful www.comnets.uni-bremen.de MAC - 11

www.comnets.uni-bremen.de MAC - 12

Assumptions for the Evaluation of Collision Resolution Methods (2) No loss: sender transmits packet until the transmission is successful Station is backlogged, as long as the transmission of a packet is repeated No buffering assumption: stations in a backlogged state do not generate additional packets Infinity assumption: the number m of stations are assumed to be infinite Probability bilit for generating a packet at a specific station in time T: P{Generate a Packet in time T}= (Poisson Distribution) T m e T m www.comnets.uni-bremen.de MAC - 13

Pure Aloha (unslotted) Basic random access method Station sends packet immediately after arrival Packet is received successfully when no other station sends during the same time Acknowledgement by receiver immediately after transmission Collision: two or more stations sending during the same time, no acknowledgement Sender sends packet again after a random time until packet is successfully transmitted (Acknowledgement) www.comnets.uni-bremen.de MAC - 14

www.comnets.uni-bremen.de MAC - 15

Analysis of pure Aloha (1) Packet generation rate [packets/s] Packet transmission time X Maximum packet transmission rate X Offered traffic Channel access rate (incl. repetitions) g [packets/s] Ratio to maximum rate G = g / g X Rate of successful channel accesses s [packets/s] Throughput S = s / s s X Infinitely many stations in the network m Simplification: Poisson distribution of the number of channel accesses of newly generated packets and repetitions: g www.comnets.uni-bremen.de MAC - 16

www.comnets.uni-bremen.de MAC - 17

Analysis of pure Aloha (2) Average number of channel accesses until packet was transmitted successfully? P{successful access}=s/g Successful Access X X X Collision X Channel access: Poisson distribution ib ti Packet interarrival time: neg. exponential distribution t www.comnets.uni-bremen.de MAC - 18

Formulas for pure Aloha P P gt T X ge dt X e G success P T X P T X before after 2G e S Ge 2G S max 1 2e 0.184 max 18.4% for G 0.5 www.comnets.uni-bremen.de MAC - 19

www.comnets.uni-bremen.de MAC - 20

S Throughput Pure Aloha protocol 0.2 1 0.18 2e 0.16 0.14 0.12 0.1 0.08 006 0.06 0.04 002 0.02 0 0 0.5 1 1.5 2 2.5 3 G G stab G lab www.comnets.uni-bremen.de MAC - 21

Slotted Aloha Receiver defines pattern of time slots of length X for the channel access Stations access the channel synchronously to the predefined slot pattern Otherwise: see pure Aloha Synchronization leads to complete overlay of packets sent in the same time slot Maximum Throughput relative to Pure Aloha? Maximum Throughput is doubled! www.comnets.uni-bremen.de MAC - 22

www.comnets.uni-bremen.de MAC - 23

Formulas Slotted Aloha P{1 access}=p{1 X}=gXe -gx Rate of successfully used time slots: P{1 access} = s = S or P{1 access} = S with = 1 / X this results in S=Ge -G S max =1/e=0.368 for G=1 www.comnets.uni-bremen.de MAC - 24

www.comnets.uni-bremen.de MAC - 25

Throughput Slotted Aloha Protocol S 0.4 0.35 03 0.3 0.25 1 e 0.2 0.15 0.1 005 0.05 0 0 0.5 1 1.5 2 2.5 3 G www.comnets.uni-bremen.de MAC - 26

Strategies for Collision Resolution Binary Exponential Backoff Channel Access Probability q r After j erroneous accesses s q r is set to 2 -j Stabilization Goal: Region of G=1 with high h throughput h t S or other more complex methods see Pseudo-Bayes- Algorithm [Bossert99] Max. Throughput hp < 0.587 Tree Method: Throughput < 0.46 www.comnets.uni-bremen.de MAC - 27

Tree Algorithm (1) Throughput < 0.43 A,B,C A,C Ø A,C C A B B A,C B A B B B 1 2 3 4 5 6 7 Time slots of the collision resolution period (Kollisionsauflösungsperiode) Example for Tree Algorithm www.comnets.uni-bremen.de MAC - 28

Tree Algorithm (2) Throughput < 0.46 M 0 (5) M 1 (5) M 0 (6) M 0 (3) M 1 (3) M 0 (4) M 0 (2) M 1 (2) M 2 (3) M 1 (4) M 2 (5) M 1 (6) M 0 (7) M 0 (1) 1 2 3 4 5 6 7 Time Slots of the Collision Resolution Period Presentation of subsets as trees. www.comnets.uni-bremen.de MAC - 29

FCFS-Algorithm (1) Throughput < 0.4878 two new packets time slot time Example for the procedure of the FCFS-Algorithm www.comnets.uni-bremen.de MAC - 30

FCFS-Algorithm (2) L1 No Access Collision L2 No Access Collision L3 R0 No Collision R1 R2 R3 State t diagram of the FCFS Algorithm www.comnets.uni-bremen.de MAC - 31

Carrier- Sensing- Methods Prerequisite: all stations can hear each other Common Topology: Bus Collision Avoidance through so-called Carrier Sensing (Abhören des Mediums auf Trägersignal) CSMA = Carrier r Sense Multiple Access Polite discussion (höflich geführte Diskussion) Persistence: Persistent t CSMA: channel is constantly monitored, access as soon as channel is free Non-persistent CSMA: monitor channel, if busy -> random time delay, listen to channel, access as soon as channel is free p-persistent CSMA: Mixture, listen to channel until free, access with probability bilit p, random time delay, listen to channel www.comnets.uni-bremen.de MAC - 32

Station 1 Station 2 Start of transmission vulnerable period Start of transmission End of transmission End of transmission Time = max. signal propagation delay between 2 stations L = mean packet length [bit] C = channel transmission rate [bit/s] X = mean packet transmission delay X = L / C Parameter = / X = C / L www.comnets.uni-bremen.de MAC - 33

CSMA: Throughput Non-persistent CSMA m stations States of the channel: free, successful transmission, collision k: number of stations ready to send a packet Other assumptions: see Aloha Protocol www.comnets.uni-bremen.de MAC - 34

CSMA: Throughput (1) CSMA throughput ( = 0,1) S(k) 0.55 0.5 045 0.45 S max 1 1 2 0.4 035 0.35 0.3 0.25 0.2 0.15 0.1 0 2 4 6 8 10 12 14 G(k) stab G(k) lab G(k) www.comnets.uni-bremen.de MAC - 35

1 CSMA: Throughput (2) CSMA throughput for different = 0,0001 S (k) 0.9 0.8 0.7 = 0,01 = 0,001 0.6 0.5 0.4 0.3 0.2 = 0,1 0.1 0 0 5 10 15 20 25 30 G (k) www.comnets.uni-bremen.de MAC - 36

Collision detection by sensing the carrier while sending Station 1 Station 2 Start of transmission Collision detection stop transmission i immediately free channel Start of transmission Collision detection stop transmission immediately free channel Time www.comnets.uni-bremen.de MAC - 37

Ethernet: CSMA/CD System Origin: Xerox proposal for local networks, coaxial cable, Bus, 10 Mbit/s Standard IEEE 802.3 (IEEE = Institute of Electric and Electronic Engineers) Kilometer range Collision detection depends on signal propagation, i.e. configuration of stations in the network Collision resolution: Binary-Exponential-Backoff Detailed analysis by simulation Estimation of throughput S for max. signal propagation: S 1 G 2 e G G G 1 e e www.comnets.uni-bremen.de MAC - 38

CSMA / CA (Collision Avoidance) Collision detection, e.g. in radio networks usually not possible, sent signal conceals external received signal Collision avoidance strategy: Sequence of access for all stations in the network Segment time into time slots of duration slightly larger than max. signal propagation Fixed or Random assignment of slots to stations Station is only allowed to access assigned time slot Time pattern is interrupted, as soon as a station is in sending mode Time slot is an implicit it token Example: IEEE 802.11b WLAN (installation at University of Bremen / NW1) www.comnets.uni-bremen.de MAC - 39

Polling Central assignment of sending right Formerly mainframe access for terminals Mainframe as master Terminals are polled cyclically Simple implementation Disadvantage: overhead of polling idle terminals results in considerable overhead for a large number of stations Improvement by so-called Binary-Countdown-Protocol o Sorted by address bits o Stations with highest address bit = 1 first o Repeat until station with highest address bit ready to send remains o In each step the address space is divided by 2: k = log 2 m o Fairness? z.b. durch zyklischen Tausch der Adressabfrage www.comnets.uni-bremen.de MAC - 40

Token-Passing Methods Examples: Token-Bus IEEE 802.4 Token-Ring IEEE 802.5 FDDI (Fiber Distributed Data Interface) General Token-Passing Characteristics: Scheduling Method: Station receives sending right through a so-called token Transfer of the sending right by sending the short token to the next station ti No Collisions Limit of max. time until sending right is acquired is possible (e.g. by limiting time or number of packets), suitable for real time services, priorities Router www.comnets.uni-bremen.de MAC - 41

Token-Bus (IEEE 802.4) (1) Token release after a max. of 10 ms Logical Ring implemented on bus topology Robust with regards to hardware failures Changes of the logical ring structure Deleting stations: Controlled: Station informs predecessor Uncontrolled, e.g. by error: after releasing a token, Station ti watches if successor releases token itself, otherwise a message is sent to all stations, successor (of successor) answers or logical ring is newly established. sh Adding stations: Stations regularly ask new stations to send a notification, station with highest address is included. Establishing new logical ring: if a station has not noticed any access to the bus over a sufficiently long period of time, the station defines itself as the first station and stations are added with the same procedure as above. www.comnets.uni-bremen.de MAC - 42

Token-Bus (IEEE 802.4) (2) Monitoring the token Destruction of token by transmission errors Generating new token by transmission errors Station with token fails Detection a missing i token: no channel access for a defined period of time, then establish new logical ring Existence of more than one token: station having token and detecting 2nd token destroys own token Distributed ib Monitoring, i no special stations performing central tasks Robust, not sensitive to errors www.comnets.uni-bremen.de MAC - 43

Token-Ring (IEEE 802.5) Point-to-point connections of neighboring stations Unidirectional mode of operation Data is passed on until it reaches the sender again Stations that are switched off or damaged are bridged (switch in the interface) Token: specific sequence of bits (no address, bit stuffing) is appended to data sent Deleting and adding stations automatically Lost token, duplicate token: Monitor Station (arbitrary but fixed station) generates new token after max. token rotation time Station failure after sending a packet: data circulates on the ring, monitor station sets a monitor bit in the frame header, frames with monitor bit set are deleted. www.comnets.uni-bremen.de MAC - 44

FDDI Fiber Distributed Data Interface ISO Standard ISO-9314 (ANSI) Fiber with 100 Mbit/s Max. extent 100 km Token Ring:Timed Token Protocol with priorities Timer: TRT: Token Rotation Timer THT: Token Holding Timer Late Count TTRT: Target Token Rotation Time Free-Token is copied before transmission in each station www.comnets.uni-bremen.de MAC - 45

FDDI Fiber Distributed Data Interface Standard High throughput Good integration of Ethernet and Token Ring Not developed further Only one data rate More expensive than Fast Ethernet t www.comnets.uni-bremen.de MAC - 46

Register Insertion Ring Decentralized organization Scheduling Method Network topology: see Token Ring Station inserts its own data Arriving data are stored and transmitted after sending the station s data Insertion of a register Propagation Delay (Verzögerungszeit)? Depends on how many stations insert data! www.comnets.uni-bremen.de MAC - 47

Register-Insertion-Ring Higher throughput than Token Ring Stations with a low traffic load after a station with high traffic load have a disadvantage Simple Realization Failure of a station: Contents of transit storage? (ARQ-Protocol) www.comnets.uni-bremen.de MAC - 48

DQDB (Distributed Queue Dual Bus) IEEE 802.6 Standard (1991) 155 Mbit/s, Data and Voice Scheduling Method Decentralized organization, synchronized system, priorities ATM-like cell structure with 48 byte payload (ATM: Asynchronous Transfer Mode) PLT (Payload Type) only 2 bit, SPR=Segment Segment Priority MAN = Metropolitan Area Network or HSLAN = High Speed Local Area Network 2 unidirectional busses, reverse direction of transmission, not limited in range/distance www.comnets.uni-bremen.de MAC - 49

DQDB Topology Cells Cell Generator Terminator... Terminator Cell Generator Cells www.comnets.uni-bremen.de MAC - 50

DQDB Cell Format 1 octet Bit 1: Busy Bit 2: Slot Type ACF Access Control Field Bit 3: PSR 53 octets VCI Virtual Channel Identifier 20 bit PLT SPR HEC Header Error Control Information Field 48 Octets Bit 4-5: RES Bit 6: Req 2 Bit 7: Req 1 Bit 8: Req 0 www.comnets.uni-bremen.de MAC - 51

DQDB Mode of Operation Cell Generator generates empty cells (Slots, busy bit = 0) Transmission of cells from generator to terminator Reservation Mechanisms: on the other bus, by setting the request bit of a cell Stations: Counter for request bits, a number of empty cells corresponding to this counter are not used Is DQDB fair? Example: 30 km fiber cable, 155 Mbit/s, how many cells are in transit? What is the effect of a very active node upstream? Bandwidth balancing: each node has to let a certain percentage 0 < f < 1 of empty cells pass without using them www.comnets.uni-bremen.de MAC - 52