IEEE Region 8 SIBIRCON-2010, Irkutsk Listvyanka, Russia, July 11 -- 15, 2010

335

Modeling VHF Air-to-Ground Multipath Propagation Channel and Analyzing Channel Characteristics and BER Performance Mohammad Asif Zaman, Sayed Ashraf Mamun*, Md. Gaffar, Md. Mushfiqul Alam and Md. Imran Momtaz Department of Electrical and Electronic Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh. *[email protected]

Abstract— In this paper, a VHF air-to-ground multipath propagation channel model is proposed and channel characteristics and BER performance are analyzed. A tapped-delay line filter model with time-varying coefficients is used to model the channel. The taps represent signals with delay corresponding to their relative propagation time. Each delayed signal is modulated in amplitude and phase by independent baseband random functions of time that produces Rayleigh fading. Doppler frequency shift is also incorporated in the channel model. Channel characteristics are analyzed in terms of power delay profile, Doppler frequency shift and burst error occurrence probability. Finally, BER performance of the channel is analyzed using computer simulation and effect of interleaving and error correcting code on the performance is observed.

I. INTRODUCTION Air-to-ground communication in the VHF band is widely used in civil and military avionics. The propagation channel is characterized by delay-spread, multipath fading and Doppler frequency shifts. These properties are determined by the environment and relative speed of the aircraft. Changes in the environment affect the reflection, diffraction and scattering of the signal and thus affect the channel response [1], [2]. This results in time-variant property of the channel. Multipath distortion is the main source of disturbance in VHF Digital Link (VDL) system in avionic communication channel, since the ground and aircraft antennas are both non-directional [3], [4]. The ground as well as all scattering bodies in the vicinity of the antennas is illuminated, generating multipath. Each of these reflected or scattered non-line of sight (NLOS) signals arrive at the receiver at different delays, thus creating delay-spread. The time delays are time-variant random process due to the dynamic nature of the propagation channel. Each NLOS signal also suffers from phase change caused by reflection, diffraction and scattering, which are also time-variant. The line of sight path (LOS) suffers from attenuation and delay but experiences no phase distortion. Multiple signals with independent delay, amplitude change and phase shifts results in multipath Rayleigh fading [1], [5]. A tapped-delay line filter model is one of the easiest yet efficient ways to model multipath channel [2], [6]. Each of the taps represents a NLOS signal. All the tap coefficients are time-varying in nature. The delays are assumed to

follow Gaussian distribution. The amplitude attenuation of each of the NLOS signals will not be the same, as the distances traveled by each of the reflected/scattered signals are not the same [7], [8]. So, amplitude attenuation is modeled as a Rayleigh random process [2], [9]. The phase distortion is modeled as a uniform random process [2]. A direct path with no phase shift is also incorporated in the channel model to represent the LOS signal path. Due to the relative motion of the aircraft, the channel also suffers from Doppler frequency shifts. Large Doppler shifts can cause BER flooring [5] and therefore is an important parameter of the channel. The frequency shift is assumed to be time varying. It is modeled by a Gaussian random process. Fading and Doppler shifts can cause large number of consecutive bits to be corrupted, resulting in burst errors [10]. The probability of burst error occurrence of the channel is computed and found to be considerably high. The burst errors degrade the BER performance of the system. To solve this problem, data interleaving and error correcting codes are used. The BER performance of the system with and without interleaving and error correcting coding is observed. II. CHANNEL MODEL The Channel is modeled by using a modified tappeddelay line filter model. The model is shown in Fig. 1. All the tap coefficients are assumed to time-varying random processes. There are k = 1, 2…..L taps in the model. Each of the taps represents a reflected/scattered NLOS signal. Number of NLOS paths = L is assumed along with one LOS path. For the modeling, the distance between the ground station and the aircraft is assumed to be 7.5 km which results in a delay of τdLOS = 25 µs for the LOS path. The delays of the NLOS paths, τdk are assumed be Gaussian random process with a mean of 30 µs and standard deviation of 5 µs. The amplitude attenuation of the LOS path, αLOS is easily determined from the distance between the aircraft and the ground station. The amplitude attenuation of the NLOS paths, αLk(t) is assumed to be Rayleigh random process [2], with mean value of 4 times of αLOS. The phase shift occurring for each of the NLOS signals, φk(t) is modeled as uniform random process in [– π, π] interval [2]. The LOS signal is added with the NLOS signals to model the multipath distortion.

978-1-4244-7626-8/10/$26.00 ©2010 IEEE

336

IEEE Region 8 SIBIRCON-2010, Irkutsk Listvyanka, Russia, July 11 -- 15, 2010

Fig. 2. Block diagram of Doppler frequency shift implementation. Fig. 1. Tapped delay line filter model of the air-to-ground communication channel.

White Gaussian noise is added to simulate channel noise. The power of the noise is determined by the signal to noise ratio (SNR). The output of the channel can be expressed as a sum of the delayed signals. For a modulated signal y(t) with carrier frequency fc ,

y (t ) = A cos(2π f c t + θ ) . The channel output without the Doppler shift, y/ch(t) is,

ych (t ) = α LOS (t ) cos {2π f c (t − τ dLOS )} + /

assumption that phase change can occur much more rapidly than amplitude or frequency change. III. CHANNEL CHARACTERISTICS The modeled channel is simulated using MATLAB. The carrier frequency is assumed to be 118 MHz [3]. For simulation, GMSK modulation scheme is used and standard data channel band of 25 kHz is assumed [11]. Usual Doppler shift range is about 100 – 200 Hz [11]. For simulation it is assumed to be centered around 150 Hz. From simulation, power-delay profile, Doppler shift and burst error occurrence probability of the channel is observed.

L

∑α

Lk

(t ) cos {2π f c (t − τ dk ) + φk (t )} + n(t ) .

k =1

Here, n(t) is white Gaussian noise. To complete the channel, Doppler frequency shift must be introduced. Hilbert transform is used to convert the real signal to complex signal. The Doppler shift, fd(t) is assumed to be Gaussian process with mean value of ±150 Hz. The probability of fd(t) being positive or negative is assumed to be equally likely. If we consider a simplified signal x(t) as the input of the Doppler frequency shifter shown in Fig. 2, where, x (t ) = cos(2π fc t )

and ,

Hilbert [ x (t )] = cos(2π fc t ) + j sin(2π fc t ) . Then, the output of the frequency shifter is xdopp(t),

A. Power delay profile During multipath propagation, the NLOS signals suffer from random delays which are assumed to be larger than the constant delay of the LOS signal, which is τdLOS = 25 µs. This causes delay spread. As a result, the received power is also spread in time. The received power versus time plot is known as power delay profile. To simulate the power delay profile, the channel is excited with a very short duration pulse at t = 0 and the channel output power is observed. The simulation results are shown in Fig. 3. From the results, it can be seen that at t = 25 µs there is a peak at the received power graph indicating the LOS signal. The other three peaks indicate the NLOS signals. The results are consistent with expected channel response.

xdopp (t ) = cos(2π f c t ) cos {2π f d (t )t} − sin(2π fc t ) sin {2π f d (t )t}

L

/

Lk

(t ) cos [ 2π { fc + f d (t )} (t − τ dLOS ) + φk (t ) ] + n (t )

Received Power (Watt)

ych (t ) = α LOS (t ) cos [ 2π { f c + f d (t )} (t − τ dLOS ) ]

∑α

x 10

1.2

= cos [ 2π { fc + f d (t )} t ] So, the final output of the channel can be expressed as:

+

Pow er delay profile of the channel

-20

1.4

1 0.8 0.6 0.4

k =1

The attenuation parameter αLk(t), phase shift parameter φk(t), and Doppler frequency shift parameter fd(t) are time-varying random processes. But they are assumed to remain constant for a particular duration of time. These particular time durations are taken as 0.2 ms, 1.25 µs and 20 µs respectively. The values are taken relying on the

0.2 0 0

1

2

3 time (sec)

4

5 -5

x 10

Fig. 3. Simulated power delay profile of the modeled channel.

IEEE Region 8 SIBIRCON-2010, Irkutsk Listvyanka, Russia, July 11 -- 15, 2010

337

B. Doppler frequency shift In an air-to-ground avionic communication channel, a relatively high Doppler frequency shift is expected due to the high velocity of the aircraft. The frequency shift may be positive or negative depending on the direction of motion of the aircraft relative to the ground station. For simulation, mean Doppler shift of ±150 Hz is assumed. The frequency shift varies with time due to the dynamic nature of the propagation channel. The effect of this frequency shift is that received signal is shifted from the carrier frequency to other frequencies. To simulate this, an un-modulated 118 MHz carrier is used as the input of the channel and the output is observed. Fourier transform is used to find the spectral characteristics of the channel output. The results are shown in Fig. 4. The x - axis is shifted so that the origin denotes the carrier frequency. Form the graph, it can be seen that the amplitude spectrum is shifted from the carrier frequency. The Gaussian nature of the frequency distribution can also be observed from the graph. C. Burst error characteristics Another very important characteristic of the channel is the probability of burst error occurrence. Burst errors, or corruption of large number of consecutive data bits, severely affect the performance of the communication system. Multipath fading and Doppler shifts can cause burst errors [2], [10]. The probability of burst error occurrence of the modeled channel is shown in Fig. 5. For comparison, the same results for AWGN channel is also shown in the same figure. From the results, we can see that AWGN channel suffers from random errors where as the modeled air-to-ground communication channel suffers from burst errors. The results show that for the modeled channel, about 24% of the total errors are burst errors, where 4 or more consecutive bits are corrupted. This value is only about 0.4% for the AWGN channel. So, the air-toground avionic communication channel is 60 times more likely to suffer from burst errors compared to AWGN channel. If burst error is defined as the corruption of 3 or more sequential bits, then about 34.7% of total errors are burst errors for the modeled channel. The results clearly show that the channel suffers from severe burst errors which can heavily degrade the BER performance. So, the communication system must employ interleavers and error correcting codes for satisfactory BER performance. Doppler Frequency Shift of the Received Signal 0.50 0.45

Fourier Amplitude

0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 -500

-400

-300

300 200 0 -200 Frequency , fd - fc (Hz)

400

500

Fig. 4. Doppler frequency shift of the received signal.

Fig. 5. Burst error characteristics of the channel.

IV. BIT ERROR RATE PERFORMANCE Bit error rate performance is very important for any digital communication system. Transmitter power, maximum symbol rate, maximum distance between transmitter and receiver, selection of interleaver and error correcting codes all depend on the channel BER performance. As the modeled air-to-ground communication channel suffers from severe burst errors, it is expected to have a poor BER performance even at high SNR values if no error correcting codes or interleavers are used. It is also expected that the performance can be improved by using interleavers and error correcting codes [12]. Simulation results confirm these assumptions. The BER curves of the channel are shown in Fig. 6. A very noticeable improvement in the BER performance is observed when interleaving and coding is used which ensures a successful communication link. From simulation results it seen that, when interleaver and error correcting codes are not used, the channel experiences a high error floor. The error floor is defined as the persistence of bit errors even at high SNR values [5]. As it is independent of noise power, Doppler frequency shift and multipath fading is the main cause for this error floor. However, the BER performance can be improved by employing interleavers and error correcting codes. Convolutional interleaver with cyclic error correcting code is used here. This causes an improvement in BER performance and the high error floor no longer exists. A (23, 12) Golay code is used here. The code converts blocks 12 bit input data to blocks of 23 bit coded data by adding appropriate redundancies. These redundancies make it possible to detect and correct errors in a code block. For decoding, Kasami decoding algorithm is used [12]. The Golay code can correct up to 3 error bits in a 23 bit code word. Golay codes are comparatively simple to generate and decode and does not increase the system complexity greatly. This makes it suitable for use in the air-to-ground avionic communication channel.

338

IEEE Region 8 SIBIRCON-2010, Irkutsk Listvyanka, Russia, July 11 -- 15, 2010 V. CONCLUSION An air-to-ground avionic communication channel is modeled and analyzed using computer simulation. A modified tapped delay line filter model with time-varying coefficients is used to model the channel. Power delay profile and Doppler frequency shift characteristics of the channel are observed. It is also seen that the channel suffers from burst errors. The BER performance simulation of the channel shows the presence of a high error floor. However, it is seen that with appropriate error correcting coding and interleaving, the BER performance of the air-to-ground avionic communication channel can be satisfactory.

0

10

-1

10

-2

BE R

10

-3

10

-4

10

-5

10

With interleaving and error correcting code Without interleaving and error correcting code

REFERENCES

-6

10 -30

-25

-20

-15

-10

-5

0

5

10

[1]

SNR (dB)

[2] Fig. 6. BER performance of the modeled air-to-ground channel without coding and interleaving, and with coding and interleaving.

However, Golay codes can correct random errors, but it can not correct burst errors. When more than 3 bits are corrupted in a 23 bit code word, the Golay code fails to correct the errors. As the channel suffers from burst errors, it is very likely that more than 3 bits will be corrupted in a 23 bit block. This makes the Golay error correcting code ineffective. Interleaving can be used to solve this problem [10], [13]. Interleaving is done after error correcting coding. The interleaver shuffles the coded bits over a span of block lengths. The span length is determined by the statistical nature of the burst errors. The bits are separated in time, so they are placed in different time positions. The space in between is filled by other coded bits. No time correlation exists between the symbols. A de-interleaver at the output reshuffles the coded data bits into the original order. A series of consecutive interleaved data bits can be corrupted in the channel, but after de-interleaving, those erroneous data bits will no longer be consecutive. So, the interleaver essentially converts burst errors into random errors [13]. The decoding of the error correcting code takes place after de-interleaving. This makes it possible for the Golay code to correct the errors. A convolutional interleaver is used here [13]. Convolutional interleavers take less memory and suffer from less delay compared to block interleavers. For simulation, a convolutional encoder with block length of 40 is used. Delay difference between two consecutive branches is assumed to be 5. From Fig. 6, it can be seen that after using error correcting coding and convolution interleaving, the BER performance of the communication system is improved greatly. The large improvement can be seen at moderate and high SNR values. But the coding and interleaving has no effect at very low SNR values. This implies that at extremely low SNR values, the noise is so high that the signal is corrupted beyond repair. But at normal operating SNR range, the coding and interleaving gives acceptable BER performance.

[3]

[4] [5] [6]

[7]

[8] [9] [10] [11] [12] [13]

J. D. Parsons, The Mobile Radio Propagation Channel, 2nd Edition, Wiley & Sons Ltd, 2000. John G. Proakis, and Masoud Salehi, Communication System Engineering, 2nd Edition, Pearson Prentice Hall, 2006. B. Roturier, B. Chateau, and A. Dedryvere, “A general model for VHF aeronautical multipath propagation channel”, Aeronautical Mobile Communication Panel (AMCP), WG-D/WP6, 1999. P. Hoeher, and E. Haas “Aeronautical channel modeling at VHFband”, IEEE Vehicular Technology Conference, 1999. Kamilo Feher, Wireless Digital Communications: Modulation and spread spectrum applications, Prentice-Hall, 1995. J. Sykora, “Tapped delay line model of linear randomly timevariant WSSUS channel”, Electronics Letters, Vol. 36, Issue 19, 2000. Bo Zheng, Qing-huaRen, Yun-jiang Liu, and Zhen-yong Chu, “Simulation of two V/UHF air-to-ground communication channel models”, International conference on wireless communication, networking and mobile computing, 2007. W. G. Newhall, J. H. Reed, “A geometric air-to-ground radio channel model”, MILCOM, vol. 1, 2002. Peyton Z. Peebles, Jr., Probability, Random Variables, and Random Process, 4th Edition, Tata McGraw-Hill, 2002. B. P. Lathi, Modern Digital and Analog Communication System, 3rd Edition, Oxford University Press, 1998. A. Malvern, “Improvements to VHF air-to-ground communication”, IEE Colloquium on Air-To-Ground Communications, 1997. Arnold Michelson, Error Control Techniques for Digital Communication, Wiley-Interscience, 1985. G. D. Forney, “Burst Correcting Codes for the Classic Burst Channel”, IEEE Trans. Comm., pp 772-781, Oct. 1971.

Modeling VHF Air-to-Ground Multipath Propagation ...

Finally, BER performance of the channel is analyzed using computer ... To solve this problem, data interleaving and error ..... statistical nature of the burst errors.

554KB Sizes 3 Downloads 173 Views

Recommend Documents

Modeling VHF Air-to-Ground Multipath Propagation ...
or scattered non-line of sight (NLOS) signals arrive at the receiver at different delays, thus creating delay-spread. The time delays are time-variant random process due to the dynamic nature of the propagation channel. Each. NLOS signal also suffers

vhf waki taki.pdf
Page 1 of 1. Page 1 of 1. vhf waki taki.pdf. vhf waki taki.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying vhf waki taki.pdf. Page 1 of 1.

VHF rt450manual-norsk.pdf
Page 3 of 27. VHF rt450manual-norsk.pdf. VHF rt450manual-norsk.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying VHF rt450manual-norsk.pdf.

VHF Navicom RT-450 GB.pdf
Sign in. Loading… Whoops! There was a problem loading more pages. Retrying... Whoops! There was a problem previewing this document. Retrying.

VHF Navicom RT-450 GB.pdf
Sign in. Page. 1. /. 2. Loading… Page 1 of 2. Page 1 of 2. Page 2 of 2. Page 2 of 2. VHF Navicom RT-450 GB.pdf. VHF Navicom RT-450 GB.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying VHF Navicom RT-450 GB.pdf.

Multipath Matching Pursuit - IEEE Xplore
Abstract—In this paper, we propose an algorithm referred to as multipath matching pursuit (MMP) that investigates multiple promising candidates to recover ...

Multipath Measurement in Wireless LANs
Retail Store Heavily Cluttered. [6]. (Note 1). 105. 170 ... stores and factories can have RMS delay spreads in excess of 100ns. .... FAX: +41 21 7293684. ASIA.

Pseudo 3-D modeling of trishear fault-propagation folding
Models including growth strata show that it is practically impossible to distinguish between ..... The smaller the original sphere, the more regular the result-.

SHORTWA'U'E RADIO PROPAGATION CORRELATION WITH ...
over the North Atlantic for a fire-year period, and the reiatioc position of the planets in the solar system, discloses some oerp interesting cor- relations. As a result ...

Multipath Medium Identification Using Efficient ...
proposed method leads to perfect recovery of the multipath delays from samples of the channel output at the .... We discuss this connection in more detail in the ...

High Input IP3 Mixer Enables Robust VHF Receivers, Design Note ...
Introduction. An increasing number of applications occupy the. 30MHz to 300MHz very high frequency (VHF) band. Television and radio broadcasting, navigation controls and amateur radios are a few examples. Modern RF component development is aimed at m

Broadband VHF Radiometry with a Notch-Filtered ...
Jul 24, 2015 - Connecting Network. Impedance ... System Model. Connecting. Network. Receiver. ≈ 9120 K. 39 MHz. −2.55. = +. ≈ 150 K ... Backup ...

man-9\kenwood-271a-vhf-radio.pdf
man-9\kenwood-271a-vhf-radio.pdf. man-9\kenwood-271a-vhf-radio.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying ...

QoSBeeManet: a new QoS multipath routing protocol ...
Dec 6, 2010 - and real time applications in the MANET context,. • Need of a .... Good Packet Delivery Ratio (the best is DSR due to the cache routing policy).

Implementation of DSSS Technology in Multipath ...
DSSS systems is found to deal with wireless security issues very ... efficiency is considered for future broadband wireless communication. .... rent demodulation. A short .... (a) Line of Sight Signal from 1st Finger (b) De- spreading code ... 44, No

Anytime Lifted Belief Propagation
Department of Computer Science, University of Wisconsin, Madison, USA. ‡ Computer Science Division, University of California, ... We present an algorithm,. Anytime Lifted Belief Propagation, that cor- responds to .... narrowing the bound on its bel

Label Propagation Through Linear Neighborhoods
In many practical applications of pattern classification and data mining, one often faces a lack of sufficient labeled data, since labeling often requires expensive.

Broadband VHF Radiometry with a Notch-Filtered ...
spectral density (PSD) is then given by. Sout = kGRx(GT Tant + ... [4] S. W. Ellingson, J. H. Simonetti, and C. D. Patterson, “Design and. Evaluation of an Acive ...

(Hf ⁄ Vhf ⁄ Uhf) Free Shipping & Wholesale Price.pdf
Cheap Ham Antenna Combiner ( Iso -Frequency ) Transit ... (Hf ⁄ Vhf ⁄ Uhf) Free Shipping & Wholesale Price.pdf. Cheap Ham Antenna Combiner ( Iso ...

man-41\icom-vhf-air-band-transceiver.pdf
Sign in. Loading… Whoops! There was a problem loading more pages. Retrying... Whoops! There was a problem previewing this document. Retrying.