PART-1
1. Why is the international availability of the same ISM bands important?
Computers, in contrast to, e.g., TV sets, travel around the world as laptops, PDAs etc. Customers want to use them everywhere. Thus it is very important to be able to use built-in WLAN adapters around the globe without reconfiguration and without licensing. Furthermore, it is much cheaper for WLAN manufacturers to design a single common system compared to many different systems for probably small markets.
2. Check out the strategies of different network operators while migrating towards third generation systems. Which are the reasons why a single commons system is not in sight?
Today’s GSM operators add the new 3G air interfaces of UMTS to their existing GSM/GPRS infrastructure networks. Current GSM/GPRS networks already offer packet and circuit switched data transmission following the Release 99 of UMTS. The operators have to install new radio access networks, i.e., antennas, radio network controller etc. as described in chapter 4. The situation is similar for operators using cdmaOne (IS-95) technology. However, these operators go for cdma2000 as this system allows them to reuse their already existing infrastructure. Thus, based on the separation of the mobile phone systems into (very roughly) CDMA and GSM operators will lead to two different major 3G systems, cdma2000 and UMTS (and their future releases). Right now, it does not seem that there is a place for a third 3G system. Current TDMA operators might move to EDGE enhanced systems or join the UMTS system. However, it is still open what will happen in China – the Chinese system TD-SCDMA was pushed by the government, but networks and devices are still missing. Currently, the majority of Chinese subscribers use GSM, some operators offer CDMA.
3. What are the means to mitigate narrowband interference? What is the complexity of the different solutions?
Several mechanisms exist to mitigate narrowband interference (which might be caused by other senders, too):
Dynamic Frequency Selection: Senders can sense the medium for interference and choose a frequency range with lower/no interference. HiperLAN2 and 802.11h use this scheme. Network operators can also this scheme to dynamically assign frequencies to cells in mobile phone systems. DFS has a relatively low complexity.
Frequency hopping: Slow frequency hopping (several symbols per frequency) may avoid frequencies with interference most of the time with a certain probability. This scheme may be used in GSM. Furthermore, wireless systems can use this principle for multiplexing as it is done in Bluetooth systems (still slow hopping as Bluetooth sends many symbols, indeed a whole packet, on the same frequency). Fast hopping schemes transmit a symbol over several frequencies, thus creating a spread spectrum. FH systems have medium complexity. Main topic is synchronisation of the devices.
Direct sequence spread spectrum: Data is XORed with a chipping sequence resulting in a spread signal. This is done in all CDMA systems, but also in WLANs using, e.g., Barker sequences for spreading (e.g., 802.11b). The signal is spread over a large spectrum and, thus, narrowband interference only destroys a small fraction of the signal. This scheme is very powerful, but requires more powerful receivers to extract the original signal from the mixture of spread signals.
4. What are the main reasons for using cellular systems? How is SDM typically realized and combined with FDM? How does DCA influence the frequencies available in other cells?
The main reason is the support of more users. Cellular systems reuse spectrum according to certain patterns. Each cell can support a maximum number of users. Using more cells thus results in a higher number of users per km². Additionally, using cells may support user localisation and location based services. Smaller cells also allow for less transmission power (thus less radiation!), longer runtime for mobile systems, less delay between sender and receiver. Well, the downside is the tremendous amount of money needed to set-up an infrastructure with many cells. Typically, each cell holds a certain number of frequency bands. Neighbouring cells are not allowed to use the same frequencies. According to certain patterns (7 cluster etc.) cellular systems reuse frequencies. If the system dynamically allocates frequencies depending on the current load, it can react upon sudden increase in traffic by borrowing capacity from other cells. However, the “borrowed” frequency must then be blocked in neighbouring cells.
5. How does the near/far effect influence TDMA systems? What happens in CDMA systems? What are countermeasures in TDMA systems, what about CDMA systems?
As long as a station can receive a signal and the signal arrives at the right time to hit the right time-slot it does not matter in TDMA systems if terminals are far or near. In TDMA systems terminals measure the signal strength and the distance between sender and receiver. The terminals then adapt transmission power and send signals in advance depending on the distance to the receiver. Terminals in CDMA systems have to adapt their transmission power very often (e.g., 1500 times per second in UMTS) so that all signals received, e.g., at a base station, have almost the same strength. Without this one signal could drown others as the signals are not separated in time.
6. Explain the term interference in the space, time, frequency, and code domain. What are countermeasures in SDMA, TDMA, FDMA, and CDMA systems?
Interference and countermeasures in:
SDMA: Interference happens if senders are too close to each other. Terminals or base stations have to keep a minimum distance. TDMA: Interference happens if senders transmit data at the same time. Countermeasures are tight synchronization and guard spaces (time gap between transmissions).
FDMA: Interference happens if senders transmit data at the same frequency. Thus, different frequencies have to be assigned to senders by organizations, algorithms in base stations, common frequency hopping schemes etc. Furthermore, guard bands between used frequency bands try to avoid interference.
CDMA: Interference happens if senders transmit data using non-orthogonal codes, i.e., the correlation is not zero. Thus, senders should use orthogonal or quasi-orthogonal codes.
PART-2
1. What characteristics do the different orbits have? What are their pros and cons?
Characteristics, pros/cons of different orbits (see chapter 5 for further figures i.e. if u want…):
• GEO: Satellites seem to be pinned to the sky; pros: fixed antennas possible, wide area coverage, simpler system design; cons: long delays, high transmission power, low system capacity (difficult SDM), weak signals at high latitudes, and crowded positions over the equator.
• LEO: low orbiting satellites; pros: low delay, lower transmission power, intersatellite routing; cons: high complexity, high system cost.
• MEO: somewhere in-between GEO and MEO
• HEO: non-circular orbits; pros: higher capacity over certain points; cons: complex systems
2. Why is there hardly any space in space for GEOs?
In order to stay synchronous with the earth’s rotation, GEOs have to use the common orbit at 35786 km. Furthermore, the inclination must be 0°. This leads to satellites stringed on this orbit like stones on a thread. Additionally, the satellites should orbit above populated regions. Thus, areas above the equator looking towards Europe, America, Asia etc. are crowded. This is also the reason why all satellites must spare some propellant to catapult them out of the orbit after their lifetime. They must not block their position.
3. What are the general problems of satellite signals travelling from a satellite to a receiver?
Attenuation caused by the atmosphere, dust, rain, fog, snow, … Blocking of signals due to obstacles (buildings, mountains). The lower the elevation the longer is the way for the signals through the atmosphere. Without beam forming high output power is needed.
4. 2G and 3G systems can both transfer data. Compare these approaches with DAB/DVB and list reasons for and against the use of DAB/DVB.
DAB and DVB both offer much higher data rates compared to 2G/3G networks. But they operate only unidirectional and bandwidth is shared (well, the capacity of a 2G/3G cell is shared, too). Thus, broadcast systems are good for distributing mass data relevant to many (in the best case all) users. Good examples are radio and TV, but also system updates, popular web content, news etc. Typically, it is to expensive to broadcast individual data. However, if broadcast bandwidth is available this is feasible, too. DAB/DVB can be complementary to 2G/3G systems. In particular if downloads are needed at higher relative speeds. Mobile phone systems have to lower their bandwidth dramatically at high speeds, while broadcast systems may still work at full bandwidth.
5. What multiplexing schemes are used in GSM and for what purpose? Think of other layers apart from the physical layer.
GSM uses SDM, FDM and TDM:
SDM: Operators design the cell layout, place base stations and reuse frequencies according to certain cluster patterns.
FDM: Regulation authorities assign channels to operators, operators assign channels to base stations, and base stations assign a certain channel to a terminal during data transmission.
TDM: Base stations assign a time-slot or several time-slots to a terminal for transmission.
6. Which components can perform combining/splitting at what handover situation? What is the role of the interface Iur? Why can CDMA systems offer soft handover?
The important characteristic of combining/splitting is that it is never performed inside the (traditional) core network. I.e., the MSCs do not notice anything from the new possibilities offered by CDMA (reception of data via more than one base station). Depending on the location of the handover (between two antennas at the same Node B, between two Node Bs, between two RNCs) the Node Bs or the serving RNC have to perform splitting/combination. The interface Iur is needed to transfer data between RNCs for combination/splitting without any interaction with the CN. For CDMA receiving signals from different base stations looks like multipath propagation. The rake receivers can thus handle both. The handover is then as soft as a change in the strongest signals in a multipath scenario. TDMA/FDMA systems like GSM cannot do this because the currently used time-slot and/or frequency may not be available in the next cell.
PART-3
1. What are the basic differences between wireless WANs and WLANs, and what are the common features? Consider mode of operation, administration, frequencies, capabilities of nodes, services, national/international regulations.
Differences:
coverage (GSM 70km cells, WLAN 100m),
data rates (GSM 50 kbit/s, WLAN 50 Mbit/s),
quality of service (WWAN voice/data rate, WLAN none/some with HiperLAN2),
transmission power (powerful base stations for WWANs, some hundred mW for WLANs),
operation (WWAN licensed, WLAN license exempt),
administration (WWAN public operators, WLAN private),
frequencies (WWAN many different national frequencies, WLAN almost common international ISM bands).
Common characteristics: similar propagation characteristics, similar problems.
2. What are advantages and problems of forwarding mechanisms in Bluetooth networks regarding security, power saving, and network stability?
Forwarding data in Bluetooth between piconets require a node jumping back and forth between these piconets. This also requires authentication in both networks, nodes that are (almost) always active and synchronous clocks if the master jumps into another piconet. If the master jumps away all network traffic in the piconet stops, all slaves have to wait until the master returns. All hopping sequences must stay synchronous during that time. Up to now not many devices are capable of forming scatternets with nodes jumping back and forth.
4. List the entities of mobile IP and describe data transfer from a mobile node to a fixed node and vice versa. Why and where is encapsulation needed?
Two kinds of entities comprise a Mobile IP implementation:
- A home agent stores information about mobile nodes whose permanent home address is in the home agent's network.
- A foreign agent stores information about mobile nodes visiting its network. Foreign agents also advertise care-of addresses, which are used by Mobile IP. If there is no foreign agent in the host network, the mobile device has to take care of getting an address and advertising that address by its own means.
A node wanting to communicate with the mobile node uses the permanent home address of the mobile node as the destination address to send packets to. Because the home address logically belongs to the network associated with the home agent, normal IP routing mechanisms forward these packets to the home agent.
Encapsulation is required between the HA and the COA, which could be located at an FA or at the MN. This is needed to make mobility transparent – the inner data packet should not notice data transfer through the tunnel, thus TTL remains untouched.
5. In what situations can collisions occur in all three networks? Distinguish between collisions on PHY and MAC layer. How do the three wireless networks try to solve the collisions or minimize the probability of collisions?
During polling, there are no collisions on the MAC layers of HiperLAN2 and Bluetooth as the access point/master controls the medium. However, in order to access the access point, nodes may transmit during a random access phase in HiperLAN2 (random channel with feedback from the access point). At this point collisions may occur on the MAC layer. For 802.11 collisions on the MAC layer are nothing unusual. The MAC algorithm with back-off solves this problem. Collisions on the PHY layer may occur in Bluetooth only if another piconet randomly jumps to the same frequency at the same time. This will destroy data for this time-slot. In HiperLAN2 different networks are separated in frequency, thus there should be not collisions besides the above mentioned during the random access phase. In 802.11 networks MAC collisions are also collisions at the PHY layer. Important packets in 802.11 have higher priorities implemented via shorter waiting times (SIFS, PIFS).
6. What are general problems of mobile IP regarding security and support of quality of service?
Mobile IP does not increase security compared to IP, on the contrary. The only additional security related function is the authentication of MN and HA. However, if MN and HA, together, want to attack an FA, nothing can prevent them. Firewalls and mobile IP do not really go together. Either reverse tunnelling or tunnelling in general drills a hole in the firewall or MNs can not operate in foreign networks. The firewall has to be integrated into the security solution. IP does not support QoS. If QoS supporting approaches like DiffServ or IntServ are used, new functions are needed for mobile IP to support QoS during and after handover. Furthermore, packets requiring certain QoS must be treated according to these requirements also inside the tunnel.
PART-4
1. Assume a fixed internet connection with a round trip time of 20 ms and an error rate of 10–10. Calculate the upper bound on TCP’s bandwidth for a maximum segment size of 1,000 byte. Now two different wireless access networks are added. A WLAN with 2 ms additional one-way delay and an error rate of 10–3, and a GPRS network with an additional RTT of 2 s and an error rate of 10–7. Redo the calculation ignoring the fixed network’s error rate. Compare these results with the ones derived from the second formula (use RTO = 5 RTT). Why are some results not realistic?
First of all the tricky part. Error rates on links, as given in the question, are always bit error rates. Under the assumption that these errors are independent (and only under this assumption!), the packet loss probability p used in the formulae can be calculated as: p = 1 - ((1-bit error rate)packet size). Using this formula, you can calculate the packet loss rates (this ignores all FEC and ARQ efforts!).
• Fixed network: BER = 10-10, MSS = 1000 byte = 8000 bit, thus the packet loss rate p = 1 - ((1-10-10)8000) ≈ 8*10-7. RTT = 20 ms: Using the simple formula, this yields a max. bandwidth of 0.93 * 8000 / (0.02 * √(8*10-7)) bit/s ≈ 416 Mbit/s.
• WLAN: The same calculation with the WLAN error rate 10-3 and additional 2 ms delay results in a packet loss rate of 0.99966 and a bandwidth of 0.93 * 8000 / (0.022 * √0.99966) bit/s ≈ 338 kbit/s. This is a good example showing why big packets cause problems in WLANs – and why FEC/ARQ is definitively needed… Real life throughput in WLANs is about 6 Mbit/s for 802.11b WLANs (if there are no other users).
• GPRS: Using GPRS with an additional 2 s RTT and a BER of 10-7 (i.e., a packet loss rate of 8*10-4) results in only 0.93 * 8000 / (2.02 * √(8*10-4)) bit/s ≈ 130 kbit/s. Well, currently GPRS offers only 50 kbit/s, but that is a limitation the simple formula does not take into account.
• In practice, the performance depends very much on the error correction capabilities of the underlying layers. If FEC and ARQ on layer 2 do a good job, TCP will not notice much from the higher error rate. However, the delay introduced by ARQ and interleaving will decrease bandwidth. Additionally, the
slow start mechanisms must be considered for short living connections. Nevertheless, it is easy to see from these simple calculations that offering higher data rates, e.g., for GPRS, does not necessarily result in higher data rates for a customer using TCP.
2. Name the advantages and disadvantages of user acknowledgements in WTP. What are typical applications for both cases?
Advantage: users can control the acknowledgement process, users may want to know if something went wrong, sometimes it is also possible to slow down a sender by inserting artificial delays in the acknowledgement process, the acknowledgement of a user is ―stronger‖ as it shows the sender that the intended receiver and not the WTP process actually got the message. Disadvantages: users have to interact, this may take some more time. Classical transactional services typically benefit from user acknowledgements, for most push service user acknowledgements are not necessary, still WTP acknowledgements can improve reliability.
3. Which properties of HTTP waste bandwidth? What is the additional problem using HTTP/1.0 together with TCP? How does HTTP/1.1 improve the situation?
HTTP is text oriented and human readable. This makes parsing for humans quite simple, however, it wastes bandwidth compared to binary representations. Using HTTP/1.0 additionally wastes bandwidth as each request uses a separate TCP connection. This requires connection setup, data transfer, and connection release for each simple element on a web page. HTTP/1.1 uses persistent connections, i.e., one TCP connection can transfer several requests.
4. Now show the required steps during handover for a solution with a PEP. What are the state and function of foreign agents, home agents, correspondent host, mobile host, PEP and care-of-address before, during, and after handover? What information has to be transferred to which entity to maintain consistency for the TCP connection?
Compare with figure 9.2. FA, CN, HA, MH should work as Mobile IP specifies. Without any PEP TCP would experience packet loss due to the change of the subnet if the old FA does not forward packets. If PEPs are used the old PEP must transfer the whole state (buffers for retransmissions, sockets, …) to the new PEP. The CN and the MH should not notice the existence of PEPs. One place to put a PEP is the FA. However, the PEP could also be located at the edge of the fixed network. PEPs work on layer 4 (in this example), while the Mobile IP components work on layer 3 - they might interact, but they do not have to.
PART-5
1. Why, typically, is digital modulation not enough for radio transmission? What are general goals for digital modulation? What are typical schemes?
Worldwide regulation always uses FDM for separating different systems (TV, WLAN, radio, satellite, …). Thus, all radio systems must modulate the digital signal onto a carrier frequency using analogue modulation. The most prominent system is the traditional radio: all music and voice use frequencies between, e.g., 10 Hz and 22 kHz. However, many different radio stations want to transmit at the same time. Therefore, all the original signals (which use the same frequency range) must be modulated onto different carrier frequencies. Other motivations for digital modulation are antenna and medium characteristics. Important characteristics for digital modulation are spectral efficiency, power efficiency and robustness. Typical schemes are ASK, PSK, FSK.
2. What is the basic prerequisite for applying FDMA? How does this factor increase complexity compared to TDMA systems? How is MAC distributed if we consider the whole frequency space as presented in chapter 1?
Modulation – Transmitters must shift all baseband signals to a carrier frequency. This is typically an analogue process and requires analogue components. Classical receivers also need filters for receiving signals at certain frequencies. Depending on the carrier frequency different antennas may be needed. Pure TDMA systems stay on one frequency, all receivers can wait on the same frequency for data. In FDMA systems receivers have to scan different carrier frequencies before they can receive signals. MAC is performed on many different layers. The WRCs (World Radio Conferences) are used for worldwide frequency assignments such as the 2 GHz range for IMT-2000. ITU controls worldwide frequency usage. National authorities regulate frequencies in different nations. On the next lower layers network operators perform MAC: frequencies usage is controlled by network planning and current load. Finally, base stations in mobile phone systems assign frequencies to terminals depending on the current availability. In WLANs network administration assigns frequencies thus forming cells.
3. Why are so many different identifiers/addresses (e.g., MSISDN, TMSI, IMSI) needed in GSM? Give reasons and distinguish between user-related and systemrelated identifiers.
Users of the GSM systems work with telephone numbers. That is all users should see. These phone numbers are completely independent of the current location of the user. The system itself needs some additional information; however, it must not reveal the identity of users. The international identification of users is done with the IMSI (=country code + network code + subscriber ID). During operation within a location area, only a temporary identifier, the TMSI is needed. This hides the identity of a user. The TMSI is not forwarded to the HLR. These are already some examples for identifiers; however, GSM provides some more:
IMEI: MS identification (like a serial number); consists of type approval code (centrally assigned), final assembly code, serial number, and spare (all three manufacturer assigned).
IMSI: Subscriber identification, stored in the SIM. Consists of the mobile country code (3 digits, e.g., 262 for Germany), the mobile network code (2 digits, e.g., 01 for the German T-Mobile), and the mobile subscriber identification number (10 digits). The mobile network code together with the mobile subscriber identification number forms the national mobile subscriber identity.
MSISDN: Mobile subscriber ISDN Number, i.e., the phone number, assigned to a subscriber, not a telephone! The MSISDN is public, not the IMSI nor the mapping MSISDN-IMSI. An MSISDN consists of the country code (up to 3 digits, e.g., 49 for Germany), the national destination code (typically 2 or 3 digits), and the subscriber number (up to 10 digits).
MSRN: The mobile station roaming number is a temporarily assigned, location based ISDN number. The VLR assigns MSRNs and forwards them to the HLR/GMSC for call forwarding.
Assignment happens either upon entering a new LA or upon request from the HLR (call-setup).
LAI: The location area identity describes the LA of a network operator. It consists of a country code (3 digits), a mobile network code (2 digits), and a location area code (16 bit). The LAI is broadcasted on the BCCH for LA identification.
TMSI: The VLR currently responsible for an MS can assign a 32 bit temporary mobile subscriber identity to an MS with a SIM. The tuple (TMSI, LAI) uniquely identifies a subscriber. Thus, for ongoing communication IMSI is replaced by (TMSI, LAI).
LMSI: An additional local mobile subscriber identity (32 bit) can be used by the VLR/HLR for fast subscriber look-up.
CI: Within a LA each cell has a unique cell identifier (16 bit). Thus, the tuple (LAI, CI) uniquely identifies a cell worldwide (global cell identity).
BSIC: The base transceiver station identity code identifies base stations (6 bit) and consists of a 3 bit network colour code and a 3 bit base transceiver station colour code.
All MSCs, VLRs and HLRs have unique ISDN numbers for identification.
4. Explain packet flow if two mobile nodes communicate and both are in foreign networks. What additional routes do packets take if reverse tunneling is required?
If MNa and MNb are both in foreign networks attached to FAa and FAb the packet flow is as follows. MNa sends packets to MNb via the Internet to HAb (actually, MNa sends to MNb’s address, the packets are only intercepted by HAb). HAb encapsulates the packets to FAb, which then forwards the packets to MNb. If reverse tunnelling is required, the packet flow is as follows: MNa sends its packets via FAa through the reverse tunnel via HAa and the Internet to HAb. HAb then forwards the packets through the tunnel to FAb, which in turn forwards the packets to MNb.
5. Compare the presented protocol stacks for WAP 2.0 and give application examples.
See figure 10.38.
WAP 1.x stack: This stack supports all ―classical‖ WAP phones and applications. As discussed in this section, there are many reasons for session services and more efficient transactional services. Thus, this stack will remain a useful part of WAP.
WAP with profiled TCP: This i-mode like scenario offers optimised HTTP and TCP. This might be more efficient then using ―pure‖ Internet solutions, but reqires changes in HTTP and TCP – and the proxy to translate.
WAP with TLS tunnelling: If end-to-end security is a must, the architecture must not break the connection. Therefore this stack offers end-to-end TLS, but can still benefit from an optimised TCP.
WAP direct: If the devices are powerful enough and delays are not too high, the standard Internet protocol stack can be used. This would be the simplest solution, which does not require any special WAP protocols any more. While the devices might be powerful enough in the future, the delay problems will remain.
