1 NeXt generation/ dynamic spectrum access/ cognitive radio wireless networks : A survey Ian F. Akyildiz, Won-Yeol Lee, Mehmet C. Vuran, and Shantidev.

1

NeXt generation/dynamic spectrum access/cognitive radio wireless networks : A survey Ian F. Akyildiz, Won-Yeol Lee, Mehmet C. Vura

n, and Shantidev MohantyGeorgia Institute of Technology

Computer Networks 50 (2006)

2

Outline

Introduction Cognitive radio The xG network architecture Spectrum sensing Spectrum management Spectrum mobility Spectrum sharing Upper layer issues Cross-layer designs Conclusions

3

Introduction

Today’s wireless networks are regulated by a fixed spectrum assignment policy the spectrum is regulated by governmental

agencies; the spectrum is assignment to license holders or

services on a long term basis for large geographical regions.

According to FCC, temporal and geographical variations in the utilization of the assigned spectrum range from 15% to 85%.

4

Spectrum Usage

The signal strength distribution over a large portion of the wireless spectrum

5

Introduction (cont’d)

Problem of the fixed spectrum management Some bands were allocated to services which have not

been utilized at all, but it has been just left unused over a decade (e.g. ERMES paging system, TFTS in-flight phone)

Unbalanced allocation due to miss-prediction of the demand (e.g. limited band for 3G system)

Difficulty for new applications/services to gain access The limited available spectrum and the inefficiency

in the spectrum usage necessitate a new communication paradigm to exploit the existing wireless spectrum opportunistically.

6

Introduction (cont’d)

Dynamic spectrum access is proposed to solve the spectrum inefficiency problems.

DARPAs approach on Dynamic Spectrum Access network, the so-called NeXt Generation (xG) program aims to implement the policy based intelligent radios know as cognitive radios. The inefficient usage of the existing spectrum can be impro

ved through opportunistic access to the licensed bands without interfering with the existing users.

7

Introduction (cont’d)

The key enabling technology of xG networks is the cognitive radio (CR). Cognitive radio techniques provide the capability to use or

share the spectrum in an opportunistic manner. Dynamic spectrum access techniques allow the cognitive r

adio to operate in the best available channel.

8

Introduction – Main Functions of CR (cont’d) The main functions for cognitive radios in xG networ

ks: Spectrum sensing

Detecting unused spectrum and sharing the spectrum without harmful interference with other users

Spectrum management Capturing the best available spectrum to meet user communic

ation requirements Spectrum mobility

Maintaining seamless communication requirements during the transition to better spectrum

Spectrum sharing Providing the fair spectrum scheduling method among coexisti

ng xG users

9

Introduction – xG Network Communication Functionalities (cont’d)

10

Cognitive Radio

A “Cognitive Radio” is a radio that can change its transmitter parameters based on interaction with the environment in which it operates.* Cognitive capability

to capture or sense the information from its radio environment

to identify the portions of the spectrum that are unused at a specific time or location

Reconfigurability The CR can be programmed to transmit and receive on

a variety of frequencies and to use different transmission access technologies by its hardware design.

* FCC, ET Docket No 03-222 Notice of proposed rule making and order, Dec. 2003

11

Cognitive Radio (cont’d)

The CR enables the usage of temporally unused spectrum, which is referred to as spectrum hole or white space.

12

Cognitive Radio - Physical Architecture (cont’d)

In the RF front-end, the received signal is amplified, mixed and A/D converted.

In the baseband processing unit, the signal is modulated/demodulated and encoded/decoded.

13

Cognitive Radio - Physical Architecture (cont’d)

The novel characteristic of CR transceiver is a wideband sensing capability of the RF front-end. RF hardware should be capable of tuning to any part of a

large range of frequency spectrum.

14

Cognitive Radio – Key Challenge of Physical Architecture (cont’d)

Limitations The wideband RF antenna receives signals from various tr

ansmitters operating at different power levels, bandwidths, and locations.

The RF front-end should have the capability to detect a weak signal in a large dynamic range.

The capability requires a multi-GHz speed A/D converter with high resolution, which might be infeasible.

Solutions Reduction of dynamic range of the signal, e.g., tunable notc

h filters Multiple antennas such that signal filtering is performed in t

he spatial domain rather than frequency domain, e.g., beamforming.

15

Cognitive Radio – Cognitive Capability Cognitive Cycle

16

Cognitive Radio – Reconfigurability Reconfigurability

is the capability of adjusting operating parameters for the transmission on the fly without any modifications on the hardware components. Operating frequency Modulation

Reconfigure the modulation scheme adaptive to the users requirements and channel conditions.

Transmission power If higher power operation is not necessary, the CR reduces the tra

nsmitter power to a lower level to allow more users to share the spectrum and to decrease the interference

Communication technology

17

The xG Network Architecture [5]

18

The xG Network Architecture- Primary Network An existing network infrastructure is generally

referred to as the primary network, which has an exclusive right to a certain spectrum band. Primary user Primary base-station

The primary base-station does not have any xG capability for sharing spectrum with xG users.

The primary base-station may be requested to have both legacy and xG protocols for the primary network access of xG users.

19

The xG Network Architecture- xG Network xG network (cognitive radio network, Dynamic Spect

rum Access network, secondary network, unlicensed network) does not have license to operate in a desired band. The spectrum access is allowed only in an opportunistic ma

nner. xG users xG base-station

provides single hop connection to xG users without spectrum access license

Spectrum broker can be connected to each network and can serve as a spectru

m information manager to enable coexistence of multiple xG networks

20

The xG Network Arch.- Access Types xG network access

xG users can access their own xG base-station both on licensed and unlicensed spectrum bands.

xG ad hoc access xG users can communicate with other xG users th

rough ad hoc connection on both licensed and unlicensed spectrum bands.

Primary network access The xG users can also access the primary base-st

ation through the licensed band.

21

xG Network on Licensed Band

xG networks is deployed to exploit the spectrum holes through cognitive communication techniques.

22

xG Network on Licensed Band (cont’d)

The challenges is due to the existence of the primary users. the detection of the presence of primary users the interference avoidance with primary users

The channel capacity if the spectrum holes depends on the interference at the nearby primary users.

spectrum handoff If primary users appear in the spectrum band occupied b

y xG users, xG users should vacate the current spectrum band and move to the new available spectrum immediately.

23

xG Network on Unlicensed Band Open spectrum policy has caused an impressive var

iety of important technologies and innovative uses. However, due to the interference among multiple heterogen

eous network, the spectrum efficiency of ISM band is decreasing.

xG networks can be designed for operation on unlicensed bands such that the efficiency is improved in this portion of spectrum. Intelligent spectrum sharing algorithm can improve the effici

ency of spectrum usage and support high QoS.

24

xG Network on Unlicensed Band (cont’d) xG uses focus on detecting the transmissions of other xG use

rs. All xG users have the same right to access the spectrum

No spectrum handoff is triggered by the appearance of other primary users

If multiple xG network operators reside in the same unlicensed band, fair spectrum sharing among these networks is also required.

25

xG Network Applications

Leased network The primary network can provide a leased network by

allowing opportunistic access to its licensed spectrum with the agreement with a third party without sacrificing the service quality of the primary users.

e.g., Mobile Virtual Network Operator (MVNO) Cognitive mesh network

xG networks have the ability to add temporary or permanent spectrum to the infrastructure links used for relaying in case of high traffic load.

Emergency network Military network

26

The xG Network Architecture - Examples Spectrum Pooling [61][62] CORVUS (Cognitive Radio approach for usage of Virtual

Unlicensed Spectrum) [8][14] exploit unoccupied licensed bands

27

The xG Network Architecture – Examples (cont’d)

IEEE 802.22 (Wireless Regional Area Networks) The first worldwide standard based on the cognitive radio

technology. Focus on constructing fixed point-to-multipoint WRAN that

will utilize UHF/VHF TV bands between 54 and 862 MHz. Specific TV channels as well as guard bands will be used

for communication in IEEE 802.22.

28

The xG Network Architecture – Examples (cont’d)

DIMSUMnet (Dynamic Intelligent Management of Spectrum for Ubiquitous Mobile-access network) [10][35] argued a case for coordinated, real-time dynamic spectrum acce

ss instead of opportunistic, uncoordinated methods common in ad-hoc military applications.

Recent work focuses on the spectrum pricing and allocation functions for spectrum brokers. [11]

29

The xG Network Architecture – Examples (cont’d)

DRiVE project (Dynamic Radio for IP Services in Vehicular Environments) [75] focus on dynamic spectrum allocation in heterogeneous net

work (broadcast technologies and cellular system) by assuming a common coordinated channel.

OverDRiVE (Spectrum Efficient Uni- and Multicast Services Over Dynamic Radio Networks in Vehicular Environments) [26] aims at UMTS enhancements and coordination of existing r

atio networks into a hybrid network to ensure spectrum efficient provision of mobile multimedia service.

30

The xG Network Architecture – Examples (cont’d)

Nautilus [73][74][15] is designed to emphasize distributed coordination

enabled spectrum sharing, without relying on centralized control.

OCRA network (OFDM-based Cognitive Radio) [5] introduces multi-spectrum transport techniques to

exploit the available but non-contiguous wireless spectrum for high communications.

31

Spectrum Sensing

The most efficient way to detect spectrum holes is to detect the primary users that are receiving data within th

e communication range of an xG user. In reality, however, it is difficult for a cognitive radio t

o have a direct measurement of a channel between a primary receiver and a transmitter.

Thus, the most recent work focuses on primary transmitter detection based on local observations of xG users.

32

Classification of Spectrum Sensing Techniques

Transmitter detection approach is based on the detection of the weak signal from a primary

transmitter through the local observations of xG users. Basic hypothesis the AWGN

the amplitude gain of the channeltransmitted signal of the primary users

33

Transmitter Detection Problem Transmitter detection problem

Receiver uncertainty (a) Shadowing uncertainty (b)

34

Cooperated Spectrum Sensing Cooperated spectrum sensing methods where infor

mation from multiple xG users are incorporated for primary user detection. allow to mitigate the multi-path fading and shadowing effect

s, which improves the detection probability in a heavily shadowed environment.

The primary receiver uncertainty problem caused by the lack of the primary receiver

location knowledge is still unsolved.

35

Interference-based Detection

The interference temperature model [21] shows the signal of a radio station designed to operate in a range at which the received power approaches the level of the noise floor.

36

Spectrum Sensing Challenges Interference temperature measurement

There exists no practical way for a CR to measure or estimate the interference temperature at nearby primary receivers. Primary receivers are usually passive devices

Spectrum sensing in multi-user networks Current interference model do not consider the effect of mu

ltiple xG users Detection capability

Detect the primary users in a very short time.

37

Spectrum Management

Since xG networks should decide on the best spectrum band to meet the QoS requirements over all spectrum bands, new spectrum management functions are required for xG n

etworks considering the dynamic spectrum characteristics

Functions of spectrum management Spectrum sensing Spectrum analysis Spectrum decision

38

Spectrum Analysis

The available spectrum holes show different characteristics which vary over time.

Spectrum analysis enables the characterization of different spectrum bands, which can be exploited to get the spectrum band appropriat

e to the user requirements. In order to describe the dynamic nature of xG netwo

rks, each spectrum hole should be characterized considering not only time-varying radio environment and but also the primary user activity and the spectrum band inf

ormation.

39

Spectrum Analysis – Parameters Interference

From the amount of the interference at the primary receiver, the permission power of an xG user can be derived, which is used for the estimation of the channel capacity

Path loss The path loss increases as the operating frequency increas

es. Therefore, if the transmission power of an xG user remains

the same, the its transmission range decreases at higher frequencies.

Wireless link errors Depending on the modulation scheme and the interference

level of the spectrum band

40

Spectrum Analysis – Parameters (cont’d) Link layer delay

To address different path loss, wireless link error, and interference, different types of link layer protocols are required at different spectrum bands. results in different link layer packet transmission delay

Holding time refers to the expected time duration that the xG user can oc

cupy a licensed band before getting interrupted. The longer the holding time, the better the quality would be.

41

Spectrum Analysis – Capacity Estimation Usually SNR at the receiver has been used for the capacity es

timation. Since SNR considers only local observations of xG users, it is not

enough to avoid interference at the primary users.

Spectrum characterization is focus on the capacity estimation based on the interference at the licensed receivers. Interference temperature model

A complete analysis and modeling of spectrum in xG networks is yet to be developed.

42

Spectrum Decision

Once all available spectrum bands are characterized, appropriate operating spectrum band should be s

elected for the current transmission considering the QoS requirements and the spectrum characteristics

43

Spectrum Management Challenges Decision model

how to combine these spectrum characterization parameters for the spectrum decision model

Multiple spectrum band decision The multi-spectrum transmission shows less quality

degradation during the spectrum handoff. Transmission in multiple spectrum bands allows lower

power to be used in each spectrum band. As a result, less interference with primary users is achieved.

how to determine the number of spectrum bands and how to select the set of appropriate bands

44

Spectrum Management Challenges (cont’d) Cooperation with reconfiguration

The CR technology enables the transmission parameters of a radio to be reconfigured for optimal operation in a certain spectrum band.

For example, if SNR is fixed, the bet error rate can be adjusted to maintain the channel capacity by exploiting adaptive modulation techniques.

45

Spectrum Management Challenges (cont’d) Spectrum decision over heterogeneous spectru

m bands In licensed bands

Consider the activities of primary users in spectrum analysis and decision in order not to influence the primary users transmission.

In unlicensed bands All the xG users have the same access rights, sophisticated s

pectrum sharing techniques are necessary.

46

Spectrum Mobility

xG networks target to use the spectrum in a dynamic manner by allowing CR to operate in the best available frequency band.

Spectrum mobility is defined as the process when an xG users changes its frequency of operation. Spectrum mobility arises when current channel conditions b

ecome worse or a primary user appears.

47

Spectrum Mobility (cont’d)

Each time an xG user changes its frequency of operation, the network protocols are going to shift from one mode of operation to another.

The purpose of spectrum mobility management in xG networks is to make sure that such transitions are made smoothly and as

soon as possible The applications running on an xG users perceive minimum

performance degradation during a spectrum handoff.

48

Spectrum Mobility Challenges Algorithms are required to decide the best available

spectrum based on the channel characteristics of the available spectrum and the QoS requirements of the applications.

Design new mobility and connection management approaches to reduce delay and loss during spectrum handoff.

Novel algorithms are required to ensure that applications do not suffer from severe performance degradation during the transitions.

49

Spectrum Mobility Challenges (cont’d)

Inter-cell handoff and vertical handoff Spectrum mobility in time domain

The available channels change over time, enabling QoS in this environment is challenging.

Spectrum mobility in space The available bands also changes as a user move

s from one place to another. Continuous allocation of spectrum is a major chall

enge.

50

Spectrum Sharing

Spectrum sharing can be regarded to be similar to generic medium access control (MAC) problems in the existing systems.

The coexistence with licensed users and the wide range of available spectrum are two of the main reasons fro the unique challenges.

51

Spectrum Sharing Process

Spectrum sensing Spectrum allocation

The allocation not only depends on spectrum availability, but it is also determined based on internal (and possible external) policies.

Spectrum access The access should be coordinated in order to prevent

multiple users colliding in overlapping portions of the spectrum.

Transmitter-receiver handshake Spectrum mobility

52

Classification of Spectrum Sharing

underlayoverlay

53

Inter-network and Intra-network Spectrum Sharing

54

Inter-network Spectrum Sharing Centralized approaches

Common Spectrum Coordination Channel (CSCC) etiquette protocol [33] for coexistence of IEEE 802.11b and 802.16a

Spectrum policy server [32] Each operator bids for the spectrum indicating the cost it will pay for t

he duration of the usage. The SPS then allocates the spectrum by maximizing its profit from th

ese bids Distributed approaches

Distributed QoS based Dynamic Channel Reservation (D-QDCR) [43] A base station of a WISP competes with its interfere BSs according t

o the QoS requirements of its users to allocate a portion of the spectrum.

55

Intra-network Spectrum Sharing Cooperative approaches

Local bargaining (LB) [15] to ensure a minimum spectrum allocation to each users and

hence focuses on fairness of users Dynamic open spectrum sharing MAC (DOSS-MAC)[40]

When a node is using a specific data channel for communication, both the transmitter and the receiver send a busy tone signal through the associated busy tone channel.

…… Non-cooperative approaches

Device centric spectrum management (DCSM) [73] The communication overhead is minimized by providing five

different system rules for spectrum allocation.

56

Spectrum Sharing Challenges Common control channel

CCC facilitates many spectrum sharing functionalities Transmitter receiver handshake Communication with a central entity Sensing information exchange

A fixed CCC is infeasible in xG networks When a primary user chooses a channel, this channel h

as to be vacated without interfering.

57

Spectrum Sharing Challenges (cont’d)

Dynamic radio range Radio range changes with operating frequency due to

attenuation variation. When a large portion of the wireless spectrum is

considered, the neighbors of a node may change as the operating frequency changes.

Control channels in the lower portions of the spectrum where the transmission

range will be higher Data channels

in the higher portions of the spectrum where a localized operation can be utilized with minimized interference

58

Spectrum Sharing Challenges (cont’d) Spectrum unit

Time dimensional The time required to

transfer information Rate dimensional

The data rate of the network Multi-code or Multi-

channel Power/code dimensional

The energy consumed for transmitting information throughput the network

three-dimensional resource-space

59

Upper Layer Issues - Routing

Common control channel Intermittent connectivity

In xG networks, the reachable neighbors of a node may change rapidly. The available spectrum may change or vanish as licensed use

rs exploit the networks Once a node selects a channel for communication, it is no lon

ger reachable through other channels The connectivity concept used for wireless networks depen

ds on the spectrum. Re-routing Queue Management

60

Upper Layer Issues – Transport Layer The performance of TCP depends on the packet loss probability

and round trip time. Wireless errors and the packet loss probability depends on

the access technology the frequency in use interference level the available bandwidth

RTT of a TCP connection depends on the frequency of operation

packet retransmissions due to higher frame error rate at particular frequency bands

spectrum handoff latency the interference level the medium access control protocol

61

Cross-layer Designs

62

Conclusions

xG networks are being developed to solve current wireless network problems resulting from the limited available spectrum the inefficiency in the spectrum usage

xG networks, equipped with the intrinsic capabilities of the cognitive radio, will provide an ultimate spectrum-aware communication paradigm in wireless communications.

63

Resources

Published Special Issues Mobile Networks and Applications, Aug. 2006 IEEE Communications Magazine,

May 2007, Cognitive Radios for Dynamic Spectrum Access Apr. 2008, Cognitive Radio Communications and Networks

IEEE Wireless Communications Aug. 2007, Cognitive Wireless Networks

IEEE Journal on Selected Areas in Communications Jan. 2008, Cognitive Radio: Theory and Application

Major Conferences IEEE International Symposium on Dynamic Spectrum Access Networks (Dy

SPAN) 2005,2007, 2008

International Conference on Cognitive Radio Oriented Wireless Networks and Communications (CROWNCOM) 2006, 2007, 2008